Internet/Tablet Analysis

Description

read the attached PDF articles 1. Internet access is a necessity: a latent class analysis of COVID-19 related challenges and the role of technology use among rural community residents. 2. The potential and prerequisites of effective tablet integration in rural Kenya
Complete the attached BLANK journal club template for EACH article by answering the questions. There is blue text to guide the response as well as a summary at the end of the worksheet.
Include a 7-9 sentences summary of EACH article at the end of the worksheet. Limit it to only one quote from each article, the rest is a summary in your own words. Read the article and summarize the findings – include elements from your journal club responses above to formulate the summary.(2022) 22:845
Dow‑Fleisner et al. BMC Public Health
https://doi.org/10.1186/s12889-022-13254-1
Open Access
RESEARCH
Internet access is a necessity: a latent
class analysis of COVID‑19 related challenges
and the role of technology use among rural
community residents
Sarah J. Dow‑Fleisner1*, Cherisse L. Seaton2, Eric Li3, Katrina Plamondon2, Nelly Oelke2,4,5, Donna Kurtz2,
Charlotte Jones6, Leanne M. Currie7, Barb Pesut2, Khalad Hasan8 and Kathy L. Rush2
Abstract
Background: Rural and remote communities faced unique access challenges to essential services such as healthcare
and highspeed infrastructure pre-COVID, which have been amplified by the pandemic. This study examined patterns
of COVID-related challenges and the use of technology among rural-living individuals during the first wave of the
COVID-19 pandemic.
Methods: A sample of 279 rural residents completed an online survey about the impact of COVID-related chal‑
lenges and the role of technology use. Latent class analysis was used to generate subgroups reflecting the patterns of
COVID-related challenges. Differences in group membership were examined based on age, gender, education, race/
ethnicity, and living situation. Finally, thematic analysis of open-ended qualitative responses was conducted to further
contextualize the challenges experienced by rural-living residents.
Results: Four distinct COVID challenge impact subgroups were identified: 1) Social challenges (35%), 2) Social and
Health challenges (31%), 3) Social and Financial challenges (14%), and 4) Social, Health, Financial, and Daily Living
challenges (19%). Older adults were more likely to be in the Social challenges or Social and Health challenges groups
as compared to young adults who were more likely to be in the Social, Health, Financial, and Daily Living challenges
group. Additionally, although participants were using technology more frequently during the COVID-19 pandemic to
address challenges, they were also reporting issues with quality and connectivity as a significant barrier.
Conclusions: These analyses found four different patterns of impact related to social, health, financial, and daily living
challenges in the context of COVID. Social needs were evident across the four groups; however, we also found nearly
1 in 5 rural-living individuals were impacted by an array of challenges. Access to reliable internet and devices has the
potential to support individuals to manage these challenges.
Keywords: Technology, COVID-19, Challenges, Internet use, Latent class analysis
*Correspondence: sarah.dow-fleisner@ubc.ca
1
School of Social Work and Centre for the Study of Services to Children
and Families, University of British Columbia Okanagan, Kelowna, BC V1V
1V7, Canada
Full list of author information is available at the end of the article
Background
COVID-19 was declared a pandemic and global health
event by the World Health Organization on March 11,
2020 [1]. The pandemic, and response to it, has had farreaching effects on many spheres of life worldwide. Globally, the economy has been impacted [2], with significant
© The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visithttp://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​
mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Dow‑Fleisner et al. BMC Public Health
(2022) 22:845
interruptions to supply chains for many products [3,
4]. Grocery supply chains had to respond to increasing
demand as consumers prepared more meals at home following restaurant closures and the demand for grocery
delivery and pickup services increased [5]. Additionally,
physical distancing measures to mitigate the spread of
the virus impacted mental health worldwide with high
levels of both depression (24%) and anxiety (21.3%)
globally [6]. In healthcare, non-urgent procedures were
postponed and non-COVID visits decreased, resulting
in unmet health needs and delayed care [7, 8]. Finally,
the pandemic led to a greater reliance on technology to
access healthcare services, connect socially, and maintain
access to basic daily needs [9, 10].
With much of healthcare and technology being centralized in urban areas, rural and remote communities
continued to face disparities in access to essential services. Systemic differences or disparities in health and
health outcomes, known as health inequities, are caused
by the unfair distribution of resources, wealth, ongoing colonialism and structural racism, and power within
and between societies (Commission of the Pan American Health Organization on Equity and Inequalities in
the Americas, 2019). Prior to the COVID-19 pandemic,
rural-urban health disparities were well documented
[11]. Rural and remote living Canadians were already disproportionately affected by environmental, social, and
economic factors, such as limited access to healthcare,
education and income opportunities, and food security
[12]. Rural residents also face greater challenges related
to COVID-19, as they are on average older, more likely to
have underlying health conditions, and have less access to
healthcare [13]. The pandemic amplified existing inequities, particularly for Indigenous Peoples, women experiencing domestic violence, people with disabilities, people
needing medical treatment, the elderly, and people in
need of housing or facing food insecurity [14]. Rurality
is yet another factor contributing to amplified inequities
during the pandemic.
Indeed, the pandemic forced many activities of daily life
to move to online modalities [15], exacerbating the wellknown rural-urban digital divide in Canada [16]. Shortly
after the pandemic started (July 2020), urban internet
speeds increased nearly 25 megabits per second (Mbps),
while rural internet speeds plateaued at 5.5 Mbps [17]. As
public health safety measures for the pandemic focused
on encouraging people to go online for work, information, essential services (e.g., food, shopping, healthcare),
and social connections, rural communities faced significant barriers compared to their urban counterparts. The
lack of equitable internet speed meant this shift was more
difficult among rural communities. Given these challenges, coupled with limited access to the internet, the
Page 2 of 11
COVID-19 pandemic has likely increased the burden of
health inequity although the extent is unknown. Thus, it
is important to better understand the multifaceted challenges that rural-living community members are facing
during COVID-19, and to explore the role of technology
use related to those challenges. Gaining such an understanding will provide the basis for better addressing the
needs and health inequities in rural-living communities.
The aims of this study were to examine the impact of
COVID-19 related challenges among rural community
members and explore differences based on sociodemographic characteristics, as well as to examine the use of
technology related to these challenges.
Methods
Study design, setting, and recruitment
This study used a cross-sectional online survey with
both quantitative and qualitative questions related to
the impact of COVID-19 related challenges and the use
of technology. Participants were eligible to complete the
survey if they were 19 years of age or older and were living in a community in a Western Canadian province considered to be rural or remote (i.e., outside the commuting
distance of a larger centre with population greater than
12,000).1 Three quarters of the region is mountainous
and the geographic area includes forests, lakes, grass
plains and deserts along with 40,000 islands [18]. Online
surveys were completed between May 29, 2020 and July
8, 2020. This survey immediately followed the first wave
of the COVID-19 pandemic in the province. During
the time of the survey, the province was in initial stages
of re-opening (first provincial re-start began mid-May
2020) [19, 20]; however, immediately prior to the survey several restrictions were in place beginning March
2020: Non-essential services, dine-in restaurants, and
parks/playgrounds were closed; non-urgent surgeries
were postponed; non essential travel was restricted; and
schools were closed and children and youth were learning from home [21].
Recruitment targeted the interior region of the province, though participation was open to all rural community members in the province. Recruitment involved
social media posts on Kijiji (Canadian Craig’s list), Facebook, and Twitter, rural community association newsletters, advertisements on rural websites, word of mouth,
and emails sent through researchers’ community networks. Social media posts targeted local rural community
pages (e.g., “Everything [community name]”) together
totally over 35,000 members and were shared through
1
Our definition of rural was adapted from Statistics Canada’s “rural and small
town” definition18 [22].
Dow‑Fleisner et al. BMC Public Health
(2022) 22:845
rural association social media feeds, as well as through 2
paid Facebook advertisements (“post boosts”) targeting
adults living within a 25 miles radius of several rural communities in the interior region of the province. Although
we were unable to track how many potential respondents
were reached in total, Kijiji ads were viewed by 21 participants, and the two Facebook advertisements had a
combined estimated audience reach of over 7400 adults.
Three $100 and one $400 draw prize incentives were
offered to promote participation. All participants provided informed consent online prior to completing the
survey. Ethics approval was received from The University
of British Columbia – Okanagan Behavioural Research
Ethics Board. All methods were carried out in accordance
with relevant guidelines and regulations.
Measures
The online survey included items related to the impact
of COVID-related challenges experienced, technology
use and challenges, and sociodemographic characteristics. The survey included both Likert-type questions and
open-ended responses. See supplemental materials for all
survey questions used in this study (Additional file 1).
COVID‑related challenges
A list of 12 challenges was generated based on the Canadian Pandemic Influenza Preparedness Guidelines [12].
The challenges were related to the impact of limitations
in four areas: social, healthcare, financial, and daily living
needs in the context of the COVID-19 pandemic. Participants rated the impact of each COVID-19 challenge
on a 5-point Likert scale (1 = not at all; 5 = extremely).
Cronbach’s alpha for the challenges scale was 0.85. For
analyses, items were dichotomized to reflect low impact
(0 = not at all, very little, somewhat) and high impact
(1 = quite a lot, extremely). Participants were also invited
to provide an open-ended response to the question “Can
you please tell us about the most significant challenge
you have faced during the COVID-19 pandemic?”
Technology access and use, positive experiences,
and challenges
Drawn from the 2018 Canadian Internet Use Survey [23]
and Statista Research Department [24], participants were
asked if they had access to internet in their home and if
they had enough connected devices to meet their needs.
Frequency of technology use was measured by asking
participants how often technology was used to connect
with others and to gather information prior to and during COVID-19. Participants were also invited to provide
open-ended responses to the following questions: “What
has been positive about your experience using technology during the COVID-19 pandemic?” and “What has
Page 3 of 11
been your biggest challenge around using technology
during the COVID-19 pandemic?”
Sociodemographic characteristics
Finally, the survey included self-reported demographic
items related to age, gender, ethnicity/race, occupation,
and education level. See Table 1 for full break down of
demographic characteristics. In addition, information
about living situation (i.e., type of home, rent or own, and
number of individuals co-habiting) and participant location was obtained. Lastly, participants were asked if they
identified as a person with a disability and about their
general health status. Self-reported sociodemographic
data were aggregated based on the variability observed in
responses. Due to lack of variability in some responses,
ethnicity/race variable was categorized into three groups:
Indigenous (e.g., First Nation, Métis, Multiracial with
Indigenous heritage), Caucasian only (e.g., no other background identified), and all other responses (e.g., Asian,
African). Education was coded into three categories: at
least some high school, a college/trade certification, and
university degree. Finally, gender-based analyses were
conducted for males and females, as there were fewer
than 5 non-binary/gender fluid respondents.
Analytic approach
Descriptive statistics
Descriptive statistics (frequencies and means/SDs) were
used to summarize the data. Initial descriptive and bivariate analyses were also conducted to examine the distribution of the challenge items and to ensure that the items
were statistically, as well as conceptually, related.
Latent class analysis
Latent class analysis (LCA) was used to generate classes,
or groups, based on similar patterns of COVID-related
impacts across 12 challenges. LCA is a finite mixturemodeling approach designed to detect latent classes,
or groups, of individuals based on a pattern of similar responses across a set of categorical indicators [25].
Using full information maximum likelihood with robust
standard error, this approach can handle missing data as
part of the response pattern. Thus, individuals were only
excluded from analyses if data were missing on all 12
indicators (n = 1).
LCA models using the dichotomized 12 challenge
items were estimated for up to a 5-class solution. Fit was
determined using three comparative fit indices: Akaike
information criterion (AIC), Bayesian information criterion (BIC), and sample size adjusted Bayesian information criterion (ABIC), as well as the bootstrap likelihood
ratio (BLRT). Lower AIC, BIC, and ABIC suggests better
fit and a non-significance likelihood-ratio test indicates
Dow‑Fleisner et al. BMC Public Health
(2022) 22:845
Page 4 of 11
Table 1 Characteristics of study population
Sample Characteristics
n
%
Age (years)
19–35
57
20.4%
36–54
95
34.1%
55+
Prefer not to answer/Missing
103
36.9%
24
8.6%
Gender
that model with an additional class (e.g., 4 vs. 5 classes)
does not offer a better fit [26, 27]. We also examined
entropy, which indicates the overall accuracy of classification, with a value closer to 1 meaning greater precision in classification of individuals into a subgroup [28].
Lastly, models were selected based on interpretability,
using item response probabilities to characterize and
name the classes to reflect the pattern of responses [25].
LCA was conducted using MPlus 7.4 [29].
Male
72
25.8%
Female
197
70.6%
Sociodemographic differences in group membership
10
3.6%
Pearson chi-square tests were used to examine whether
differences in challenge group membership were related
to categorical sociodemographic variables. In variables
where cell sizes were small, Fisher’s exact tests were used
to estimate significance. Bivariate analyses with LCA
classes were conducted in Stata 15/MP [30].
Other/Preferred not to answer/Missing
Race/Ethnicity
Caucasian
210
75.3%
Indigenous
36
12.9%
Other
Prefer not to answer/Missing
27
9.7%
6
2.2%
Disability
Yes
37
13.3%
No
232
83.1%
10
3.6%
19.0%
Prefer not to answer/Missing
Education
At least some high school
53
Trades certification/diploma
124
44.4%
University degree
101
36.2
Missing
1
0.4%
Number of children (0–18) in the home
None
50
17.9%
1
32
11.5%
2
32
11.5%
3 or more
20
7.2%
Missing
145
51.9%
Number of adults (19–64) in the home
None
36
12.9%
1
118
42.3%
2
45
16.1%
3 or more
21
7.5%
Missing
59
21.2%
Number of older adults (65+) in the home
None
49
17.6%
1
58
20.8%
2
12
4.3%
Missing
160
57.3%
Home type
Single-family home
168
60.2%
Home on a farm/ranch
65
23.3%
Multifamily home (apartment, townhouse, condo)
46
16.5%
Rent or Own Home
Rent
65
23.3%
Own
204
73.1%
Missing
10
3.6%
Thematic analysis of open‑ended questions
A trained research assistant coded the open-ended
responses in NVivo 12 (qualitative data analysis software) using qualitative thematic analysis [31]. Following
close reading of the open-ended responses, main themes
were identified to develop categories. Data coded to
each category were analyzed inductively to identify patterns in semantic content, develop a thematic summary
of the data, and select quotes to illustrate key findings
[32]. Coded data were carefully reviewed by two research
team members (EL and KR), and emerging themes were
discussed and refined with the research team. The qualitative analyses were completed alongside the quantitative
results to expand and enrich the description of rural residents’ experiences related to the pandemic.
Results
Sample descriptive statistics
Surveys were completed by 279 participants (70.6%
female), ranging in age from 18 to 85 (M = 49.5,
SD = 14.6). Participants identified their race/ethnicity as
12.9% Indigenous, 75.3% Caucasian, and 9.7% other (e.g.,
Asian, south Asian, African Canadian), with 2.2% missing data. The vast majority of participants (273, 98.2%)
reported having access to the internet at home, consistent with use of an online survey. Among this highly connected sample, 243 participants (87.1%), responded they
had enough connected devices to meet their household
needs. Most participants used computers (242, 86.7%)
and smartphones (242, 86.7%), followed by tablets (157,
56.3%), voice-assistant systems (30, 10.8%), and other
devices (24, 8.6%) to connect to the internet. The majority of participants (198, 71%) reported an increase in frequency of technology use to connect with family/friends
and 145 (51.9%) increased frequency of technology use
Dow‑Fleisner et al. BMC Public Health
(2022) 22:845
Page 5 of 11
Table 2 Participant ratings of impact of challenges faced during COVID-19
Total n
% reporting
high impact
Social Needs
Limited access to family/friends
279
78.1
Limited ability to provide support to others
277
76.1
Healthcare Needs
Limited access to healthcare services (e.g., hospital, doctor)
271
55.9
Limited access to mental health services
214
46.9
Limited access to social /support groups (e.g., addiction groups)
208
56.5
Limited access to public health information
266
20.8
Financial Needs
Limited income opportunities
235
56.0
Challenges paying my bills/rent/mortgage
275
31.0
Daily Living Needs
Limited access to daily necessities (e.g., food, water)
278
23.8
Limited access to options for food/grocery shopping
279
40.6
Limited access to stable internet/mobile connection
275
31.8
Limited access to childcare
128
33.9
Responses were dichotomized to indicate low impact (not at all, very little, somewhat) and high impact (quite a lot, extremely)
Table 3 Model fit information for 1 to 5 class LCA models
Classes
AIC
BIC
ABIC
BLRT
Entropy
Smallest
class N
1-class
3684.54
3728.07
3690.04


2-class
3284.65
3375.34
3296.07
0.807
100
3-class
3233.55
3371.40
3250.90
− 1830.27***
0.739
75
4-class
3199.21
3384.22
3222.50
5-class
3176.59
3408.76
3205.82
− 1617.33***
− 1578.77***
− 1548.61***
0.821
40
0.829
32
AIC Akaike’s information criterion, BIC Bayesian information criterion, ABIC Sample size adjusted Bayesian information criterion, BLRT Bootstrap likelihood ratio test
N = 278
***
p < 0.001 to gather information compared to before the onset of the pandemic. See Table 1 for a summary of participant characteristics. COVID challenges Overall, 78.1% of participants reported they were highly impacted by limited access to family/friends and 76.1% reported being highly impacted by limited ability to support others. Conversely, only 20.8% reported being highly impacted by a limited access to public health information and 23.8% were highly impacted by a lack of access to daily necessities. Nearly a third (31.8%) of participants also reported being impacted by a lack of access to stable internet/mobile internet. See Table 2 for the proportion of participants reporting high impact of COVID-19 related challenges. LCA results The 4-class model was the best solution that fit the data (Table 3). AIC and ABIC were all lower for the 4-class as compared to a 3-class model. Although the BIC was slightly higher for a 4-class model as compared to the 3-class model, the 4-class model had better entropy and better interpretability. We examined the pattern of responses to the challenge items across the four classes and used those to describe and assign names to the groups. Class descriptions We assigned the following names to the four groups based on the pattern of challenge impacts: 1) Social Challenges (35%), 2) Social and Health Challenges (32%), 3) Social and Financial Challenges (14%), and 4) Social, Health, Financial, and Daily living Challenges (19%). Dow‑Fleisner et al. BMC Public Health (2022) 22:845 Page 6 of 11 Table 4 Class of challenges due to limited access to resources related to social, daily living, healthcare, and financial needs Indicators Class 3 Class 4 Total Sample Social Challenges Social & Health Challenges Class 1 Class 2 Social & Financial Challenges Social, Health, Financial, and Daily Living Challenges Probability Probability Probability Probability Probability Limited access to family/friends 0.78 0.54 0.96 0.75 0.93 Limited ability to provide support to others 0.76 0.51 0.89 0.89 0.91 Limited access to healthcare services 0.56 0.25 0.68 0.43 1.00 Limited access to mental health services 0.47 0.12 0.58 0.32 1.00 Limited access to social /support groups 0.57 0.19 0.79 0.37 1.00 Limited access to public health information 0.21 0.03 0.21 0.08 0.65 Limited income opportunities 0.56 0.26 0.41 1.00 0.96 Challenges paying my bills/rent/mortgage 0.31 0.00 0.04 1.00 0.84 Social Needs Healthcare Needs Financial Needs Daily Living Needs 0.24 0.02 0.20 0.21 0.72 Limited access to options for food/grocery shopping 0.41 0.12 0.44 0.38 0.89 Limited access to stable internet/mobile con‑ nection 0.32 0.10 0.42 0.15 0.68 0.34 0.17 0.36 0.24 0.77 0.916 0.846 0.967 0.915 35% (98) 32% (88) 14% (40) 19% (52) Limited access to daily necessities Limited access to childcare Group Membership probability (γ) Percent (n) 278 Probability greater than 0.50 used to define and name groups. Bold indicates an elevated probability of challenge impact per indicator. For each indicator, a higher probability indicates a high challenge impact associated with a limited access to resources and needs The Social Challenges group had an elevated probability of experiencing high impacts related to unmet social needs, with the lowest probability of impact across daily living, healthcare, and financial needs. The Social and Health Challenges group was characterized by a high probability of experiencing impacts related to both social needs and healthcare access challenges, with lower probabilities of either financial or daily living challenges. The Social and Financial Challenges group was characterized by a high probability of both social and financial related challenge impacts, but lower probabilities of challenge impacts related to healthcare or daily living needs. Lastly, the Social, Health, Financial, and Daily Living Challenges group showed a high probability of being impacted by all COVID-related challenges examined. See Table 4 for LCA item probabilities and class membership probabilities. Association between sociodemographic characteristics, technology use, and LCA groups Sociodemographic characteristics and technology use were examined across the four groups (Table 5). Bivariate analyses revealed older adults (55+) were more likely to be in the Social challenges and Social and Health challenges groups, and less likely to be in Social and Financial and Social, Health, Financial, and Daily living challenges groups. Conversely, there was a higher percentage of young adults (19–35) in the Social, Health, Financial, and Daily living challenges and a higher percent of middleaged adults (36–54) in the Social and Financial classes than expected. There were also significant differences in class by race/ethnicity (p = 0.005) and disability status (p = 0.001). In particular, there was a higher percentage of Indigenous respondents and respondents with disabilities in the Social and Health challenges and Social, Health, Financial, and Daily living challenges groups, and a higher percentage of other non-Caucasian respondents in the Social, Health, Financial, and Daily living challenges groups. There were no statistically significant differences between class membership related to gender (p = 0.34), education (p = 0.11), home type (p = 0.51) or ownership (p = 0.13), number of children in the home (p = 0.68), or number of seniors in the home (p = 0.89). Analyses also showed an association between reporting not having enough connected devices with being in the Social, Health, Financial, and Daily living Challenges Dow‑Fleisner et al. BMC Public Health (2022) 22:845 Page 7 of 11 Table 5 Distribution of profile membership by sociodemographic characteristics and technology use Class 1 (35%) Class 2 (32%) Class 3 (14%) Class 4 (19%) Social Challenges Social & Health Challenges Social & Financial Challenges Social, Health, Financial, and Daily Living Challenges χ2 (df), sig 12.56(6), p = 0.05 Age 19–35 35.1% 24.6% 14.0% 26.3% 36–54 28.7% 29.8% 21.3% 20.2% 55+ 41.8% 35.9% 8.7% 13.6% Race/Ethnicitya Indigenous 19.4% 41.7% 8.3% 30.6% Caucasian 38.3% 31.6% 16.3% 13.9% Other responses 37.0% 14.8% 11.1% 37.0% 18.56(6), p = 0.01 Gender Male 43.7% 28.2% 9.9% 18.3% Female 32.5% 33.0% 15.2% 19.3% 3.34(3), p = 0.34 Disability Yes 13.5% 40.5% 8.1% 37.8% No 39.4% 31.2% 15.5% 13.9% 18.74(3), p < 0.001 Enough connected devicesa Yes 37.6% 31.0% 14.8% 16.5% No 18.8% 31.3% 12.5% 37.5% 9.50(3), p = 0.03 Technology use to connect with people More use 31.8% 37.4% 15.2% 15.7% Same or less use 47.7% 15.4% 13.9% 23.1% 12.46(3), p = 0.006 Technology use to get information More use 26.2% 39.3% 16.6% 17.9% Same or less use 47.5% 22.0% 12.7% 17.8% 15.02(3), p = 0.002 100% adds up across Classes. Total N = 278. Number of responses missing by sociodemographic characteristic: age (n = 24); gender (n = 10); race/ethnicity (n = 6); disability (n = 7); connected devices (n = 4) a Due to small cell size, Fisher’s exact test were used to estimate significance level group (p = 0.03). Additionally, those using technology the same or less often to connect with others or gather information during than before the pandemic were most likely to be in the Social Challenges group, whereas greater technology use to gather information was associated being in the Social and Health challenge group. Contextualizing Covid‑19 challenges Participants’ open-ended responses to the most significant challenge they faced during the COVID-19 pandemic paralleled the challenges described above. Similar to the LCA, social needs were a consistent challenge across participants. For example, family (n = 90; 32%) and social (n = 89; 32%) challenges were frequently described by participants as their greatest challenge. One rural-living participant described the difficulty supporting and being disconnected from an elderly family member: “Father in law went to the hospital with a fall and now must go in a care home because of dementia and we haven’t seen him in two months”. Challenges related to disconnection were also described by some participants as affecting their mental health. One participant, for example, wrote: “When I could not go out to see my therapist or visit with friends it made me more of a shut-in.” Job (n = 78; 28%) and financial-related (n = 25; 9%) challenges surrounding losing income opportunities and the inability to pay bills were described as the most significant challenge for many participants, while others described access to healthcare (n = 70; 25%) and daily necessities (n = 43; 15%) as their greatest challenges. Finally, technology-related challenges emerged as the most significant challenge during the pandemic for 12 participants (4%), as one participant described: “Internet was always bad, but now it’s fairly useless. With kids doing school from home, they miss out on information because of our horrible internet services”. However, the qualitative analyses also revealed resilient outcomes not seen in the quantitative analyses. In particular, the open-ended responses from 225 participants who responded to what has been most positive about their experiences using technology during COVID-19 revolved around several themes. One quarter of the Dow‑Fleisner et al. BMC Public Health (2022) 22:845 sample (n = 67; 24%) described technology as affording them greater convenience, such as access to meetings or healthcare, without having to travel. As one participant explained: “The variety of different platforms to access healthcare without having to leave the house”. A number of participants (n = 42; 15%) explained that technology supported social connection, as they stayed in touch with family and friends virtually. One participant described their family’s creative solution to social distancing: “Zoom family gatherings…even game night! Feel much closer to family I can’t see in person.” A similar number (n = 40; 14%) pointed to the use of technology to increase their knowledge, and stay informed during the pandemic. A smaller number of participants (n = 11; 4%) described the benefit of technology in terms of providing safety and options for “staying home safe” and avoiding unnecessary exposure to the virus. Finally, a few participants (n = 10; 3.6%) described their most positive experience with technology as providing leisure, or an enjoyable way to pass the time, as one participant explained: “Keeping me occupied while at home”. The use of technology to manage pandemic-related challenges was evident. However, we also found that some participants had significant challenges related to technology. The most common response to participants’ biggest challenges using technology during the COVID19 pandemic (n = 87, 31%) was technology issues/quality. Some participants experienced regular issues with internet quality, while others described how increased demand was slowing or even stopping internet service: “Our internet is so slow and capped so low, we rarely are able to complete a video call or stream a movie if its not right at the beginning of the month...even then it is slow and difficult”. Several participants (n = 18; 6.5%) pointed out financial issues, speaking to the need to purchase laptops for children now schooling at home and data limits being exceeded. One participant described having to make a trade-off in access of one life line for another, explaining that the biggest challenge to technology use was: “The need to pawn my laptop and computer for food and fuel.” A lack of digital/technology literacy meant some participants (n = 16; 5.7%) felt vulnerable as most everyday activities relied on digital technologies, as one participant described:“Sometimes feeling [I] don’t have the tech knowledge to do stuff ”. Participants (n = 16; 5.7%) also expressed various challenges that related to the sense of loneliness and isolation, explaining that technology could not replace face-to-face interactions. The reliance on technology for social connection was described as “less personal” and “depressing”. Others (n = 15; 5.4%) felt burdened by “digital fatigue” or technology overuse, as one participant described: “The burdens of increased screen time, significant eye strain and doubling up on Page 8 of 11 amount of meetings.” A minority of participants (n = 10; 3.6%) expressed concerns about safety and security, as well as misinformation on the internet. Others resented the fact that having the internet was an expectation, as one participant reported: “I resent that you have to have the Internet. It is not a luxury, it is a utility”. Discussion The purpose of this study was to describe the challenges rural-living individuals faced during the early months of COVID-19, as well as explore technology use and challenges. Findings indicate that people living rurally in a Western Canadian province were most impacted by challenges related to unmet social needs and access to reliable internet in the first 4 months of the COVID-19 pandemic. Although limited access to family or friends and limited ability to support others were the highest rated challenges among survey participants, the LCA afforded us unique insights into four patterns of challenge impacts affecting different sub-sets of rural community members. Although all four groups experienced social challenges, importantly, 65% of participants also reported high levels of challenges related to daily, healthcare, or financial needs during the COVID-19 pandemic. Additionally, 1 in 5 of these participants indicated experiencing challenges in each of these areas. Recent research conducted in the United States reported negative impacts of the COVID-19 pandemic on unemployment, life satisfaction, and well-being in rural communities [33]. These authors reported the impacts were consistent across sex, education level, and race/ethnicity. In the present study, however, we found several sociodemographic differences in the types of challenges participants experienced. Older adults experienced more challenges related to social and health needs. Although older adults are more at risk from COVID-19 in terms of mortality and hospitalization [34], it was young adults (19–25 years of age) who were impacted by a wider variety of challenges during the pandemic, which may relate to the unique milestones associated with young adulthood. Young adults may be in the process of starting a family, be less financially secure, and in need of more supports to manage their family’s needs. Additionally, middle-aged adults (36–54 years of age) were most likely to be in the Social and Financial challenges category. During middle adulthood, individuals may be caring for children and for aging parents, which may increase the impact on financial needs. Indigenous people, those with non-white ethno-racial identities, and those with a disability were also more likely to be in the Social, Health, Financial, and Daily Living challenges category compared to their Caucasian counterparts. A double jeopardy therefore exists for Indigenous and ethnic minority groups, who are both Dow‑Fleisner et al. BMC Public Health (2022) 22:845 at greater risk of COVID-19 [35], and were also likely to experience a wider variety of challenges during the first wave of the pandemic. This is likely related to the existing inequities faced by these equity-denied groups. These findings are significant because they are among the first to examine the impacts of COVID-19 challenges among subgroups of rural-living Canadians, and they support the concept that the pandemic has amplified inequity [14]. Unprecedented measures were taken to slow the spread of the virus and flatten the curve in the first wave of the pandemic when no vaccine was available; however, these measures have had disproportionate consequences for different people in rural communities [14, 35]. Historical analyses have suggested that measures to mitigate the spread of COVID-19 will have unequal consequences and economic impacts, exacerbating heath inequities [36]. The present findings provide preliminary evidence for this suggestion in rural contexts. The rural-urban digital divide grew at the onset of COVID-19 [16], further marginalizing rural citizens. Although limited access to stable internet/mobile connection was not among the highest rated challenges, it was the most prevalent concern reported to open-ended inquire about the biggest technological challenges. It is notable that the Social, Health, Financial, and Daily Living Challenges group had the highest probability related to a lack of access to reliable internet compared to other challenge groups, suggesting that a lack of reliable internet may exacerbate other challenges. Likewise, participants’ lack of stable internet connection underscores some of the difficulty these rural-living adults had in accessing healthcare services. Indeed, COVID-19 has catalyzed a rapid massive shift to telemedicine to decrease person-to-person contact, and slow the spread of the virus [9, 37]. However, reliance on virtual connections to support healthcare has raised concerns of further health disparities and inequity for rural populations without the necessary digital infrastructure. Telemedicine used to its full capacity (e.g., video for assessment and diagnosis) requires adequate broadband access, which is often limited or unavailable in rural and underserved settings [38]. Although the vast majority of these online survey participants reported having access to the internet at home, the need and demand for internet had increased. For many, technology provided solutions to connect with family and healthcare virtually, stay informed, school children from home, and keep entertained. Yet, many rural residents that responded to this survey reported challenges related to quality, reliability, and affordability of internet and equipment even though they had access to the internet. Rural communities continue to experience complex challenges related to internet access, as Page 9 of 11 only 46% of rural communities in Canada have access to high speed internet [39]. This can undermine the use of technology as a social-distancing option during a pandemic. Access is limited by broadband capacity, but also technology literacy as well as data limits and financial concerns, particularly given that there were financial and income-related challenges. Findings echo international claims that good internet access is a social determinant of health and wellbeing [40]. Overall, findings indicate that the challenges different rural community members were experiencing during the first wave of the pandemic were multidimensional and likely further exacerbated by unequal access to reliable, high speed internet. Supporting rural communities requires interventions that address localized and systemwide challenges to access to technology and essential services. Limitations and future research As this was an online survey of rural community residents in a Western Canadian province primarily in the interior region, the results are not generalizable to those without internet access or residing in other areas. Additionally, this was a highly connected sample, with 98% reporting some access, and 100% having enough access to complete an online survey. This may not be reflective of most rural-living individuals. However, even among this highly connected sample, there were still key technological challenges that arose. We asked participants whether they had enough connected devices to meet the needs of their household to allow for subjective interpretation of what number of devices might be enough for different respondents; future research might include a standardized measure of number of devices to determine how many devices different rural-living respondents need. Another key limitation was the small sample size for comparative analyses, which impacted the ability to examine sociodemographic differences across challenge categories. These analyses should be replicated with a larger sample to ensure the accuracy and usefulness of the categories. That said, for the qualitative analyses this sample size was robust. The primarily female, Caucasian, who had completed trades or University education sample may not reflect the broader rural demographic; yet, important sociodemographic differences in challenges were evident. Samples including more people facing socioeconomic disadvantage would likely report greater challenges and impacts related to the costs and difficulty of access to technology. Future research would be useful to further explore groups and communities most at risk during a pandemic, especially those isolated, living with chronic diseases, mental health or substance challenges, and older Dow‑Fleisner et al. BMC Public Health (2022) 22:845 adults in long-term care. Although examining disaggregated data carries the risk of reinforcing stigmatization, doing so using a purposeful process with the intention of understanding and addressing inequities can be a force for positive change [41]. Page 10 of 11 Abbreviations LCA: Latent Class Analysis; AIC: Akaike’s information criterion; BIC: Bayesian information criterion; ABIC: Sample size adjusted Bayesian information crite‑ rion; BLRT: Bootstrap likelihood ratio test. Supplementary Information The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s12889-​022-​13254-1. Conclusions Our findings provide insight about the complex and diverse needs among rural-living community members during the first wave of the COVID-19 pandemic. Taken for granted everyday activities such as grocery shopping, attending in-person medical appointments, and engaging in social interactions were moved to online platforms. As the pandemic continues, the lack of reliable internet access in rural communities will further enlarge the digital divide between the rural and urban citizens, further challenging Canada’s universal healthcare system. The contribution of our present study is threefold. First, the survey illustrates varied sets of challenges (social, health, and financial) experienced by rural residents during the first wave of the COVID-19 pandemic. Study findings also highlight how technology use is connected to these challenges. Second, the LCA provides a new methodological approach in which to develop meaningful patterns based on citizens’ challenges. The new profiling technique offers opportunity to organize categories of citizens that experienced different forms of challenges during the COVID-19 pandemic. This type of categorization approach will be informative for policy-makers, decision-makers, and practitioners to further explore new technological solutions that address specific identified needs. Third and finally, the qualitative findings provide insights from the perspective of rural citizens during the COVID-19 pandemic. The emerging themes presented in this study captured a diversity of areas of concern that are worth further exploration. For instance, the inter-relationship between technology literacy and concerns about misinformation and trustworthiness of online news are compounded by accessibility and affordability of technologies and the costs of multiple electronic devices and internet subscription plans. Fear of the future and pandemic restrictions of in-person interactions are exacerbating worry that rural citizens have about their adaptability. In conclusion, this research has uncovered concerning intersections of technology use, human rights and equity, and future policy planning in the context of rural communities’ access to digital technologies. More work is required at all levels of government and health and education and workplace systems to ensure reliable internet access is affordable and available to all. Additional file 1. Rural community challenges and technology use: Questionnaire Items Acknowledgements We would like to acknowledge a student research assistant, Sara Amis, who assisted with the thematic coding of the open-ended responses. Authors’ contributions All authors contributed to the conceptualization of the project and contrib‑ uted to the study design. CS oversaw the data collection. S D-F completed the data analyses. All authors contributed to manuscript drafts and reviewed the final manuscript. The author(s) read and approved the final manuscript. Authors’ information Not applicable. Funding Funding for this project was provided by the University of British Columbia Okanagan’s Eminence Program [GR015968, 2019], the UBC Okanagan Work Study program, and the Regional Socio-Economic Development Institute of Canada Fund [62R34654, 2020]. The study design, data collection, analysis, interpretation and manuscript writing were completed independent of the study funders. Availability of data and materials The datasets this study is based on are stored on secure servers at The Uni‑ versity of British Columbia - Okanagan. Anonymized data are available upon request from the corresponding author. Declarations Ethics approval and consent to participate This study received ethics approval from at [University removed for deidenti‑ fied review] Behavioural Research Ethics Board. All participants completed online informed consent forms. All methods were carried out in accordance with relevant guidelines and regulations. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Author details 1 School of Social Work and Centre for the Study of Services to Children and Families, University of British Columbia Okanagan, Kelowna, BC V1V 1V7, Canada. 2 School of Nursing, University of British Columbia Okanagan, Kelowna, BC V1V 1V7, Canada. 3 Faculty of Management and Principal’s Research Chair (Tier 2) in Social Innovation for Health Equity and Food Security, University of British Columbia Okanagan, Kelowna, BC V1V 1V7, Canada. 4 Rural Coordination Centre of British Columbia, Vancouver, Canada. 5 Department of Community Health Sciences, Cummings School of Medi‑ cine, University of Calgary, Calgary, Canada. 6 Faculty of Medicine, University of British Columbia Okanagan, Kelowna, BC V1V 1V7, Canada. 7 School of Nursing, University of British Columbia, Vancouver, BC V6T 2B5, Canada. 8 Computer Science, University of British Columbia Okanagan, Kelowna, BC V1V 1V7, Canada. Dow‑Fleisner et al. BMC Public Health (2022) 22:845 Received: 28 May 2021 Accepted: 7 April 2022 References 1. Cucinotta D, Vanelli M. WHO declares COVID-19 a pandemic. Acta Biomed. 2020. https://​doi.​org/​10.​23750/​abm.​v91i1.​9397. 2. Gupta M, Abdelmaksoud A, Jafferany M, et al. COVID-19 and economy. Dermatol Ther. 2020. https://​doi.​org/​10.​1111/​dth.​13329. 3. Burki T. Global shortage of personal protective equipment. Lancet Infect Dis. 2020. https://​doi.​org/​10.​1016/​S1473-​3099(20)​30501-6. 4. Cohen J, Rodgers YVM. Contributing factors to personal protective equip‑ ment shortages during the COVID-19 pandemic. Prev Med. 2020. https://​ doi.​org/​10.​1016/​S0091-​7435(20)​30287-5. 5. Gray RS. Agriculture, transportation, and the COVID-19 crisis. Can J Agric Econ. 2020. https://​doi.​org/​10.​1111/​cjag.​12235. 6. Castaldelli-Maia JM, Marziali ME, Lu Z, et al. Investigating the effect of national government physical distancing measures on depression and anxiety during the COVID-19 pandemic through meta-analysis and metaregression. Psychol Med. 2021. https://​doi.​org/​10.​1017/​S0033​29172​10009​ 33. 7. Ammi M. Collateral damage: the unmet health-care needs of nonCOVID-19 patients. 2020, https://​theco​nvers​ation.​com/​colla​teral-​dam‑ age-​the-​unmet-​health-​care-​needs-​of-​non-​covid-​19-​patie​nts-​145934 (Accessed 22 Dec 2020). 8. Canadian Institute for Health Information. Overview: COVID-19’s impact on health care systems. 2021, https://​www.​cihi.​ca/​en/​covid-​19-​resou​rces/​ impact-​of-​covid-​19-​on-​canad​as-​health-​care-​syste​ms/​overv​iew-​covid-​ 19s-​impact-​on (Accessed 14 Oct 2021). 9. Monaghesh E, Hajizadeh A. The role of telehealth during COVID-19 out‑ break: a systematic review based on current evidence. BMC Public Health. 2020. https://​doi.​org/​10.​1186/​s12889-​020-​09301-4. 10. Nguyen MH, Gruber J, Fuchs J, et al. Changes in digital communication during the COVID-19 global pandemic: implications for digital inequality and future research. Soc Media Soc. 2020. https://​doi.​org/​10.​1177/​20563​ 05120​948255. 11. Desjardins PM. Regional disparities in Canada: Interprovincial or urban/ rural? Région et Développement n° 33; 2011. 12. Government of Canada. Public health measures: Canadian pandemic influenza preparedness: planning guidance for the health sector. 2018, https://​www.​canada.​ca/​en/​public-​health/​servi​ces/​flu-​influ​enza/​canad​ ian-​pande​mic-​influ​enza-​prepa​redne​ss-​plann​ing-​guida​nce-​health-​sector/​ public-​health-​measu​res.​html (Accessed 16 April 2020). 13. Henning-Smith C, Tuttle M, Kozhimannil KB. Unequal distribution of COVID-19 risk among rural residents by race and ethnicity. J Rural Health. 2021. https://​doi.​org/​10.​1111/​jrh.​12463. 14. Canadian Human Rights Commission. Statement - Inequality amplified by COVID-19 crisis. 2020, https://​www.​chrc-​ccdp.​gc.​ca/​eng/​conte​nt/​state​ ment-​inequ​ality-​ampli​fied-​covid-​19-​crisis (Accessed 22 Dec 2020). 15. Weeden SA and Kelly W. Addressing the digital divide: COVID-19 and the importance of connecting rural Canada. 2020, http://​crrf.​ca/​ri-​digit​ al-​divide/. 16. Canada West Foundation. NEWS RELEASE | COVID-19 pandemic has exac‑ erbated the digital divide between rural and urban communities. 2020, https://​cwf.​ca/​resea​rch/​publi​catio​ns/​news-​relea​se-​covid-​19-​pande​mic-​ has-​exace​rbated-​the-​digit​al-​divide-​betwe​en-​rural-​and-​urban-​commu​ nities/ (Accessed 16 Sept 2020). 17. Canadian Internet Registration Authority. New internet performance data shows urban speeds improving while rural speeds plateau. 2020, https://​ www.​cira.​ca/​newsr​oom/​state-​inter​net/​new-​inter​net-​perfo​rmance-​data-​ shows-​urban-​speeds-​impro​ving-​while-​rural (Accessed 5 Sept 2020). 18. Province of British Columbia. Geography of BC. 2021, https://​www.​welco​ mebc.​ca/​Choose-​B-C/​Explo​re-​Briti​sh-​Colum​bia/​Geogr​aphy-​of-B-​C#:​ ~:​text=​Mount​ains%​20cov​er%​2075%​20per%​20cen​t,bird%​20spe​cies%​ 20and%​20land%​20mam​mals. (Accessed 25 May 2021). 19. Province of British Columbia. BC’s restart plan: Next steps to move BC through the pandemic. 2020, https://​medic​alsta​ff.​islan​dheal​th.​ca/​sites/​ defau​lt/​files/​covid-​19/​latest-​commu​nicat​ions/​bc-​gover​nment/​bc-​resta​ rt-​plan-​next-​steps-​6may20-​bcgov.​pdf (Accessed 22 Dec 2020). Page 11 of 11 20. Province of BC. Premier outlines plan to restart B.C. safely. 2020, https://​ news.​gov.​bc.​ca/​relea​ses/​2020P​REM00​26-​000826 (Accessed 14 Oct 2021). 21. Kotyk A. Scroll through this timeline of the 1st year of COVID-19 in B.C. Van‑ couver: CTV News; 2021. https://​bc.​ctvne​ws.​ca/​scroll-​throu​gh-​this-​timel​ ine-​of-​the-​1st-​year-​of-​covid-​19-​in-b-​c-1.​52849​29 (Accessed 14 Oct 2021) 22. du Plessis V, Beshiri R, Bollma RD. Definitions of rural: research paper. Ottawa: Statistics Canada; 2002. 23. Statistics Canada. Canadian internet use survey. 2018, https://​www.​ statc​an.​gc.​ca/​eng/​stati​stical-​progr​ams/​instr​ument/​4432_​Q2_​V2#​a03 (Accessed 16 April 2020). 24. Statista. Number of internet connected and connectable devices in the United States from 2016 to 2020 2017, https://​www.​stati​sta.​com/​stati​ stics/​741735/​us-​number-​of-​inter​net-​conne​ctable-​devic​es/ (Accessed 16 April 2020). 25. Lanza ST, Cooper BR. Latent class analysis for developmental research. Child Dev Perspect. 2016. https://​doi.​org/​10.​1111/​cdep.​12163. 26. Morgan GB. Mixed mode latent class analysis: an examination of fit index performance for classification; 2015. https://​doi.​org/​10.​1080/​10705​511.​ 2014.​935751. 27. Nylund KL, Asparouhov T, Muthén BO. Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study; 2007. https://​doi.​org/​10.​1080/​10705​51070​15753​96. 28. Masyn KE. Latent class analysis and finite mixture modeling. In: Little TD, editor. The Oxford handbook of quantitative methods. New York: Oxford University Press; 2013. p. 551. 29. Muthén L, Muthén B. MPlus user’s guide. Los Angeles: Muthén & Muthén; 2012. 30. StataCorp. Stata statistical software: release 15. College Station: StataCorp LLC; 2015. 31. Elliott V. Thinking about the coding process in qualitative data analysis. Qual Rep. 2018. https://​doi.​org/​10.​46743/​2160-​3715/​2018.​3560. 32. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006. https://​doi.​org/​10.​1191/​14780​88706​qp063​oa. 33. Mueller JT, McConnell K, Burow PB, et al. Impacts of the COVID-19 pan‑ demic on rural America. Proc Natl Acad Sci U S A. 2021. https://​doi.​org/​ 10.​1073/​pnas.​20193​78118. 34. World Health Organization. WHO delivers advice and support for older people during COVID-19 2020, https://​www.​who.​int/​news-​room/​featu​ re-​stori​es/​detail/​who-​deliv​ers-​advice-​and-​suppo​rt-​for-​older-​people-​dur‑ ing-​covid-​19 (Accessed 14 Oct 2021). 35. Mein SA. COVID-19 and health disparities: the reality of "the great equalizer". J Gen Intern Med. 2020. https://​doi.​org/​10.​1007/​ s11606-​020-​05880-5. 36. Bambra C, Riordan R, Ford J, et al. The COVID-19 pandemic and health inequalities. J Epidemiol Community Health. 2020. https://​doi.​org/​10.​ 1136/​jech-​2020-​214401. 37. Smith AC, Thomas E, Snoswell CL, et al. Telehealth for global emergencies: Implications for coronavirus disease 2019 (COVID-19). J Telemed Telecare. 2020. https://​doi.​org/​10.​1177/​13576​33X20​916567. 38. Hirko KA, Kerver JM, Ford S, et al. Telehealth in response to the COVID-19 pandemic: implications for rural health disparities. J Am Med Inform Assoc. 2020. https://​doi.​org/​10.​1093/​jamia/​ocaa1​56. 39. Canadian Radio-television and Telecommunications Commission. Broad‑ band fund: closing the digital divide in Canada 2021, https://​crtc.​gc.​ca/​ eng/​inter​net/​inter​net.​htm (Accessed 21 May 2021). 40. American Medical Informatics Association. Consider broadband access a social determinant of health 2017, https://​healt​hitan​alyti​cs.​com/​news/​ amia-​consi​der-​broad​band-​access-​a-​social-​deter​minate-​of-​health. 41. British Columbia’s Office of the Human Rights Commissioner. Disaggre‑ gated demographic data collection in British Columbia: the grandmother perspective 2020, https://​bchum​anrig​hts.​ca/​wp-​conte​nt/​uploa​ds/​ BCOHRC_​Sept2​020_​Disag​grega​ted-​Data-​Report_​FINAL.​pdf (Accessed 24 May 2021). Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. BioMed Central publishes under the Creative Commons Attribution License (CCAL). Under the CCAL, authors retain copyright to the article but users are allowed to download, reprint, distribute and /or copy articles in BioMed Central journals, as long as the original work is properly cited. British Journal of Educational Technology doi:10.1111/bjet.12870 Vol 51 No 2 2020 498–514 The potential and prerequisites of effective tablet integration in rural Kenya Carolyn J. Heinrich , Jennifer Darling-Aduana and Caroline Martin Carolyn J. Heinrich, Ph.D., is the Patricia and Rodes Hart Professor of Public Policy, Education and Economics in the Peabody College at the Vanderbilt University. Her research focuses on education, workforce development, social welfare policy, program evaluation and public management in the United States and other international contexts. Jennifer Darling-Aduana, M.Ed., is a doctoral student in the Department of Leadership, Policy and Organization, Peabody College, Vanderbilt University. Her research focuses on mechanisms to reduce educational inequities in the United States, with particular attention to student–teacher–curriculum interactions and how digital learning transforms teaching and learning processes in schools. Caroline Martin, M.Ed., is a Research Analyst at the Tennessee Department of Education. She has taught and led professional development and conducted educational research in a variety of settings, including large urban school districts in the United States and in international settings such as Nicaragua and Kenya. Address for correspondence: Carolyn J. Heinrich, Patricia and Rodes Hart Professor of Public Policy and Education, Department of Leadership, Policy, and Organizations, Peabody College, 230 Appleton Place, Nashville, TN, USA. Email: carolyn.j.heinrich@vanderbilt.edu Abstract This study investigates how pedagogical, cultural and institutional factors interact with technical knowledge in educational technology integration and how they relate to equitable and effective technology use in low-resource settings. In the context of a one-to-one tablet initiative in rural Kenya, we explore how these factors constrain or support access to technology, instructor capacity, student engagement and student learning, as well as their implications for reducing educational and digital divides. We employ a mixed methods, such as a quasi-experimental (prepost, nonequivalent control group) research design that draws on data from classroom observations, teacher interviews, student surveys and focus groups, and assessments of student academic performance to generate evidence on classroom practices and student learning in schools with access to tablets, while also highlighting core challenges to successful technology integration. Our findings contribute to the identification of prerequisites and supporting factors for successful educational technology integration, as well as policy levers and school-based strategies that are likely to increase equitable access to quality learning experiences in schools in low-resource contexts. Introduction In the face of an increasingly competitive, global knowledge economy, governments, schools and nongovernmental organizations are turning to information and communication technology (ICT) as a means to increase student engagement and learning. Policymakers also see ICT as a promising strategy for improving access to educational resources and enhancing teachers’ ability to meet diverse student needs, particularly in low-resource settings where schools may lack sufficient funds to meet basic operating and educational costs that support access to the physical infrastructure, technical capacity and human capital (Herodotou, 2018; Twining, Raffaghelli, © 2019 British Educational Research Association Effective tablet integration in rural Kenya    499 Practitioner Notes What is already known about this topic • A lack of funding, planning and infrastructure hinder information and communication technology (ICT) integration. • Procurement of educational technology and infrastructure improvement efforts has reduced digital divides, but learning divides persist in implementation. • Insufficient teacher technology expertise and professional development constrain teacher’s effective use of educational technology in classrooms. What this paper adds • We employ mixed methods—triangulating the student assessment data with data from student surveys and focus groups, teacher interviews and classroom observations—to identify how pedagogical, cultural and institutional factors interact with technical knowledge in ICT integration in ways that support or constrain student learning in low-resource contexts. • We find that more attention is needed for cultural factors that interact with pedagogical and technical skills to ensure that the classroom instructors’ attention is equitability distributed in ways that discourage in-class “tracking” and differential access to quality learning experiences, such as some teachers’ disregard of “slow learners” in the classroom. • In low-resource contexts, providing even basic levels of infrastructure (eg, a consistent power source) and access to general technical knowledge requires more creative and concerted efforts from school leadership and instructors, such as the offer of tutorials and reading clubs outside of the class to expand access and improve the use of devices. Implications for practice and/or policy • We find that device sharing can have positive effects on peer-to-peer learning, which suggests that policymakers in Kenya should weigh the benefits of achieving one-toone device access against the potential advantages of alternative investments, such as expanding professional development on the integration and use of currently available devices. • Increasing opportunities for peer-to-peer learning and exchange (among teachers and students) and building shared capacities for ICT integration can help reduce technical issues and lost instructional time. Albion, & Knezek, 2013; Warschauer, Knobel, & Stone, 2004; Wong, Li, Choi, & Lee, 2008). Among these are policymakers in the Ministry of Education in Kenya, which rolled out the ICT Integration in Primary Education (or Digital Literacy) project as one of its flagship programs for improving teaching and learning in Kenya’s public primary schools. The project components include improvements in the ICT infrastructure and procurement of devices, development of digital content and capacity building of the teachers, which were built on rural electrification efforts initiated with the 2006 Energy Act. Despite the promise and hype, the literature is rife with discussions of the challenges of integrating technology and ensuring equitable access across a broad range of educational contexts © 2019 British Educational Research Association 500    British Journal of Educational Technology Vol 51 No 2 2020 (Hohlfeld Ritzhaupt, Barron, & Kemker, 2008; Warschauer & Matuchniak, 2010). In this research, we delved deeply into a setting in rural Kenya, where public schools and a community-based nonprofit partner are collaborating in implementing a one-to-one tablet initiative in primary schools under the Digital Literacy Project. The goals of this eReader (tablet) initiative, supported by the Lwala Community Alliance (LCA), include improving access to educational resources, enhancing classroom learning and increasing student achievement for students in North Kamagambo, Kenya. Toward that end, the LCA designed and implemented a pilot program that provided eReaders equipped with course books and supplementary books to Class 6 teachers and students at three primary schools in this region. At approximately five percent, the cost of laptops and tablets (eReaders) may not only be a more viable option in low-resource contexts, but they are also potentially more suitable for younger (primary school) learners (Herodotou, 2018; Tamim, Borokhovski, Pickup, & Bernard, 2015). Our study addressed the following key questions within this research context: (1) How, and to what extent, was tablet integration associated with any observed changes in students’ educational opportunities in rural Kenya? (2) What are the primary challenges to successful technology integration in resource-constrained contexts? (3) What policy levers and school-based strategies are likely to improve the equitable access to quality learning experiences and overcome persistent infrastructure challenges within this context? ICT integration in low-resource, educational contexts: Theory and evidence Theoretical framing Two of the most widely used frameworks for conceptualizing investigations of ICT integration are the substitution, augmentation, modification and redefinition (SAMR) and technological pedagogical content knowledge (TPACK) models (Mishra & Koehler, 2006; Puentedura, 2013). The SAMR (see Figure 1) provides a scaffold for characterizing learning tasks in terms of the depth and complexity of technology integration, broadly classifying the technology use as either enhancement or transformation. We situate the use of technology in North Kamagambo schools Figure 1: Substitution, augmentation, modification and redefinition (SAMR) model. Source: Puentedura (2013) [Colour figure can be viewed at wileyonlinelibrary.com] © 2019 British Educational Research Association Effective tablet integration in rural Kenya    501 primarily in the enhancement domain, in part because of the limitations of the technology (eReaders) and the ICT infrastructure (eg, lack of Internet access). In the most basic form (substitution), teachers use the technology as a substitute for previous instructional approaches, with no changes in instructional processes. To the extent that technology use improves on teaching or learning process (that could be undertaken with or without technology), the SAMR categorizes this technology use as augmentation (Puentedura, 2013). Alternatively, for technology use to facilitate transformative teaching—that substantially alters learning tasks in ways not possible without the technology—teachers are expected to use the technology for the significant redesign or reimagining of instructional approaches and learning opportunities (Puentedura, 2013). Teacher approaches to technology integration in the classroom are influenced not only by their knowledge of the technology and how to enact features embedded in it for enhancing or transforming learning (Orlikowski, 2000; Orlikowski and Gash, 1994; Rogoff, 2003), but also by their pedagogical and content knowledge, as shown in the TPACK model in Figure 2 (Mishra & Koehler, 2006). This framework identifies intersections of two or three of these domains (technological, pedagogical and content knowledge). The TPACK illuminates our understanding of how teachers conceive of and enact the technology in ways that are consistent with their pedagogical beliefs and practices and/or their content expertise (Hilton, 2016). Because the eReaders in North Kamagambo, Kenya were used to support access to books for language learning and reading comprehension, we focus primarily on the TPK (technological and pedagogical knowledge) intersection in the TPACK model. We also draw the sociocultural theory into our discussion of these domains. The sociocultural theory that submits cultural norms and conventions transacted by students and teachers in the classroom will influence how teachers and students understand the Figure 2: Technological pedagogical content knowledge (TPACK). Source: Mishra and Koehler (2006) [Colour figure can be viewed at wileyonlinelibrary.com] © 2019 British Educational Research Association 502    British Journal of Educational Technology Vol 51 No 2 2020 properties of the tablets, as well as how they rely on other individuals and classroom resources to support their learning (Nasir & Hand, 2006). For example, the TPK intersection motivates us to examine how teachers deploy their technological and pedagogical knowledge in grappling with the limitation of an insufficient number of tablets to facilitate one-to-one device access in their classrooms. The sociocultural theory, moreover, leads us to ask how in the face of higher than desired student-to-tablet ratios, cultural norms may affect student access to or interactions around shared devices. Used in combination, these theoretical frameworks help us to identify, classify and interpret the pedagogical, cultural, institutional and technical factors observed in this study of ICT integration. Evidence on ICT integration in low-resource settings Existing research confirms that a range of pedagogical, cultural, institutional and technical factors have the potential to contribute to (or reduce) ongoing inequities in the use of educational technology to support student learning (Hohlfeld et al., 2008; Warschauer & Matuchniak, 2010). For example, studies of ICT integration in low-resource settings have found more turnover and variability in teaching and administrative staff, which hinders the planning for and implementation of educational technology in classrooms (Warschauer et al., 2004). This personnel challenge also likely reduces the pool of technical knowledge available to educators in a given school, particularly in countries where the professional development on ICT integration is limited. And it may further dilute the effectiveness of technology-based initiatives in subsequent years and preclude teachers from moving beyond substitution (in the SAMR) to reimagining what is possible with technology use. Research shows that even when teachers have confidence in or experience with the technology being introduced, they are frequently challenged in lowresource contexts by disadvantages such as larger class sizes, more students with limited technology experience, and inadequate pedagogical and other instructional supports (Darling-Aduana & Heinrich, 2018; Warschauer et al., 2004). Some of the most common barriers to ICT integration identified in prior research in developing country contexts include: insufficient teacher technology expertise, ineffective educational software, access issues and lack of alignment with educational norms or expectations (BuabengAndoh, 2012; Pelgrum, 2001; Venezky, 2004). Mndzebele (2013) identified the lack of funding, planning and professional development as major obstacles to ICT implementation in Swaziland. Likewise, in Ghana, 85 percent of preservice teachers reported that they lacked appropriate training to use ICT (Gyamfi, 2016). While the lack of Internet connectivity was observed as a limiting factor across continents, the lack of reliable electricity also restricted the technology use in studies set in Africa (Kenya, South Africa) and Asia (Cambodia) (Richardson, 2011; Stols et al., 2015). Multiple studies have also shown that across contexts, access to technical support, professional development and other forms of assistance expand general technical knowledge that is foundational to the enhancement and transformation (in SAMR) and to the interaction of technological and pedagogical knowledge (TPK) in ways that support technology use (Buabeng-Andoh, 2012; Pelgrum, 2001; Richardson, 2011; Stanhope & Corn, 2014; Venezky, 2004). Through our theory-informed investigation and in-depth depiction of the educational technology integration in a rural, Kenyan community, we build on the contributions of prior research to identify some of the prerequisites for improving student outcomes through ICT integration in low-resource contexts. We also draw out new insights for educators and policymakers. Our mixed methods study goes beyond the technical challenges of ICT integration to also examine the pedagogical, cultural and institutional factors that support or constrain the effectiveness of ICT integration in increasing student learning and engagement, formed through the analysis and © 2019 British Educational Research Association Effective tablet integration in rural Kenya    503 triangulation of the assessment data, student surveys and focus groups, teacher interviews and classroom observations. We begin by describing our research setting, samples, and intervention, study data and measures and methods below. Study samples, data and methods Setting, samples and intervention The eReader initiative began in North Kamagambo, Kenya in 2016, through a collaboration between the LCA and rural, government-funded primary schools in this region of Western Kenya. The eReaders were provided by Worldreader, an international provider of tablets to developing countries, and distributed by the LCA for teacher and student use in three primary schools in the region. With the objective of understanding how the introduction of the eReaders would affect student learning, the LCA and the partnering researchers implemented a mixed methods study. A quasi-experimental, nonequivalent control group design was used in selected schools to participate. The LCA Education Team first categorized all 13 schools in the region by their average scores on the 2014 Kenya Certificate for Primary Education (KCPE) test into three distinct achievement tiers (low, middle and high). A total of 10 primary schools in North Kamagambo subsequently submitted proposals to participate in the eReader initiative. The LCA Education Team then selected two proposals from each of the three preestablished achievement tiers, while also factoring in both the treatment and control schools’ commitment to work with LCA and the intent to involve one school from each subarea. Because the selection of classrooms for eReader distribution was made via the criteria discussed above (and not via random assignment), we adjust for pretreatment differences in estimating associations between the eReader program and student’s outcomes and do not interpret any estimates as causal. Table 1 presents summary statistics and tests of statistical significance for differences between the characteristics of the treatment and comparison groups at the baseline, including pretreatment academic assessments. The additional baseline survey questions designed by the LCA were intended to gauge students’ access to books at school in the absence of eReaders, as well as at home, and to measure student motivation to read and caregiver support for reading and learning at home. These descriptive statistics show that students in classrooms receiving the eReaders scored significantly lower on three measures of academic performance at the baseline (before the 2016 school year): oral reading fluency in Kiswahili and English, and Kiswahili comprehension. For the other five measures of pronunciation and comprehension, there were no statistically significant, pretreatment differences in academic performance between the treatment and comparison group members. In addition, children in classrooms with eReaders reported having more access to books at school and at home at the baseline, but they were also older and significantly more likely to report that they “only read when they had to.” Within treatment schools, LCA distributed 150 eReaders to Class 6 classrooms in February 2016 in proportion to the number of teachers and students at each school, with the intent for each school to have a sufficient number of eReaders to realize a one-to-one ratio between students and the tablets. All eReaders were loaded with Class 6 workbooks and supplementary reading in Kiswahili and English. Teachers integrated the tablets in math, reading, social studies, science, Kiswahili and religion classes. Kiswahili, one of the official languages of Kenya, was the primary focus of instruction in 22 percent of observations. © 2019 British Educational Research Association © 2019 British Educational Research Association 0.88 0.51 44.67 68.74 2.40 3.42 3.79 2.77 0.71 0.50 12.88 0.12 0.46 0.43 0.37 0.51 0.25 0.21 95 95 95 84 95 95 95 95 94 94 93 94 94 94 93 95 95 95 Mean 0.48 0.50 0.44 0.41 0.50 0.50 0.45 0.50 0.15 0.32 0.22 0.25 23.10 22.19 1.88 2.09 2.16 1.95 Std. Dev. 128 128 128 128 127 127 127 127 127 127 128 128 128 119 128 128 128 128 N (students) 0.69 0.29 0.18 0.20 0.69 0.15 0.64 0.54 12.39 0.16 0.89 0.51 63.23 95.98 3.18 3.64 3.38 2.87 Mean Comparison group 0.47 0.46 0.39 0.40 0.46 0.36 0.48 0.50 0.10 0.37 0.19 0.21 31.68 33.92 2.04 2.05 2.06 1.96 Std. Dev. −0.32 0.22 0.07 0.01 −0.23 0.28 0.08 −0.04 0.49 −0.04 −0.009 0.006 −18.565 −27.246 −0.780 −0.220 0.407 −0.099 Mean difference (T-C) .000 .000 .188 .893 .000 .000 .243 .526 .005 .349 .736 .841 .000 .000 .004 .434 .155 .710 p-value Notes Statistically significant mean differences are shown in bold. The comparison group consists of students attending schools that volunteered for the eReader initiative but were not selected to implement it. 1Some students did not complete the English correct words per minute assessment, resulting in fewer observations for the analysis of this outcome. Analyses were conducted to assess the sensitivity of study findings to the loss of observations. Baseline academic performance % correct words: Kiswahili % correct words: English Kiswahili correct words per minute English correct words per minute1 Kiswahili comprehension English comprehension Kiswahili comprehension (incorrect) English comprehension (incorrect) Student characteristics (pretreatment) Parents are primary caregiver Male Age Access to less than five books at school Access to 5–10 books at school Access to more than 10 books at school Less than five books at home Child: I read only when I have to Caregiver rarely/never reads to child Caregiver rarely/never checks schoolwork N (students) Treatment group Table 1: Baseline characteristics of eReader treatment and comparison groups, 2016 school year 504    British Journal of Educational Technology Vol 51 No 2 2020 Effective tablet integration in rural Kenya    505 Data collection and measures Analyses of assessment and survey data were supplemented with emergent findings from a grounded theory analysis of data collected from classroom observations, student focus groups and teacher interviews. Below, we describe our data collection processes and resulting data sources in detail. Assessment data The program administrative data included baseline (pre) test scores for assessing student achievement and end line (post) test scores that enable us to examine associations between student tablet use and changes in their academic performance relative to other primary schools. Because pilot testing suggested that the fluency and comprehension levels of Class 6 students would be too advanced for the Early Grade Reading Assessment (EGRA) and Progress in International Reading Literacy Study (PIRLS) assessments, a custom evaluation tool was developed by an LCA Monitoring and Evaluation (M&E) team member. This tool draws on the EGRA and PIRLS assessments, as well as the input from a U.S.-based elementary school psychologist who regularly uses standardized assessments to evaluate the reading abilities of primary school children. The assessment consists of evaluation of reading abilities containing three subtests on pronunciation, oral reading fluency and comprehension in both Kiswahili and English, sourced from Class 8 Kiswahili and English textbooks. The scoring of student performance on the assessments was calculated individually per subtest. Pronunciation of each word was scored on a 0-1 scale, where 1 point was awarded for the correct pronunciation and 0 points otherwise. During the reading passages, the students’ total reading time and the number of incorrectly read words were tracked. Both metrics were used to calculate correct words per minute (CWPM). The comprehension section included multiple-choice questions, one-answer open-ended questions and multiple-answer open-ended questions. Student surveys A student survey was also administered to gather baseline information on student demographics, home environments, study habits, etc. (see again the measures in Table 1), as well as during the end line assessment, which included an additional set of questions to gauge students’ educational aspirations. A total of 109 students from treatment schools and 144 students from comparison schools completed the baseline academic assessments and survey in January and February of 2016. These same assessments were completed by 112 students from treatment schools and 136 students from comparison schools at the end-of-the-school year (in November 2016). After linking the baseline and end line data to the survey data, a total of 223 observations with complete records were available for analysis—95 students in the treatment group and 128 students in the comparison group—although as indicated in Table 1, some students did not complete the English CWPM assessment. Our analysis of the sensitivity of our results to the loss of observations (without English CWPM) did change any research conclusions. Classroom observations We also conducted classroom observations of tablet use in the summer of 2016 in North Kamagambo, Kenya. Across all classroom observations, we used a well-tested, research-based instrument that enables observers to record the extent to which an instructional session (and integration of educational technology) facilitates quality learning opportunities for students (Burch, Good, & Heinrich, 2016), with some minimal adaptations to account for differences in classroom language use and infrastructure in rural Kenya. The observation instrument incorporates multiple dimensions that capture aspects of the physical environment; curricular content and structure; instructional model; interactions between teachers, students and the technology; student and teacher engagement; and any assessment of learning. The ratings of digital and © 2019 British Educational Research Association 506    British Journal of Educational Technology Vol 51 No 2 2020 blended instruction are recorded on a 0-4 (5-point) scale; see additional information on each dimension in Appendix S.1 (online). Researchers also recorded the time lost to technology problems, the number of students per device, time allocated to various instructional strategies, and detailed narrative vignettes of instruction, activities and interactions in the classroom. A total of 36 classroom observations were conducted in the treatment and comparison schools. Student focus groups During the end line data collection, a random sample of students from both treatment and comparison schools participated in focus group discussions. Students were asked to provide their opinions on the use of eReaders in treatment schools and more generally on reading behaviors in treatment and comparison schools. (The full focus group protocol is available in Appendix S.2 online). A total of 17 students from treatment schools and 26 students from comparison schools participated in the focus groups. Teacher interviews The research team also conducted interviews with teachers to provide the context and insight into teachers’ experiences. The interview data were collected using a semi-structured interview protocol with interview topics, probes and both closed- and open-ended questions. The interview topics included instructor background, instructional practices, support for the tablet use, tablet access and use by student subgroups, assessment of the effectiveness of tablets in the classroom and plans for their ongoing use. (Refer to online Appendix S.3 for the full protocol). In total, eight classroom teachers were interviewed. Methods of analysis We analyzed the data both quantitatively and qualitatively, using triangulation across sources of information, classrooms and settings to confirm the validity and reliability of analytical findings. In analyzing the qualitative data, interviews and focus groups were recorded, transcribed and subsequently analyzed in conjunction with observation and the survey data on tablet use in the classroom to identify emerging themes using a grounded theory approach. Spot-checking was used to check coding consistency. We also searched for exceptions and alternative explanations to challenge preconceptions and personal biases. In regression analyses of the relationship of the tablet use to student academic outcomes, we estimated two alternative specifications: one that estimates the change in student achievement (1) from the beginning to the end of the 2016 school year (with the gain score as the dependent variable, Ait−Ait−1), and the other (2) that predicts the end line level of student achievement (Ait), controlling for the baseline student achievement (on the same measure) and other student characteristics at the baseline (Xit−1) as described in Table 1. Ait − Ait −1 = + 1 eRit + 2 Xit −1 + it (1) Ait = + 1 eRit + 2 Xit −1 + 3 Ait −1 + it (2) We estimate robust, clustered standard errors that account for student clustering within classrooms. Given the 2016 rollout of the eReader program, we only have one baseline measure of achievement. Because we observe baseline differences between students in eReader and © 2019 British Educational Research Association Effective tablet integration in rural Kenya    507 comparison classrooms (suggesting the potential for unobserved differences in student characteristics as well), we do not make any causal assertions about the relationship between eReader use (eRit) and changes in student outcomes. Nonetheless, research has established that the use of value-added models such as those that control for lagged measures of the dependent variables often substantially reduce bias in estimates (Chetty, Friedman, & Rockoff, 2014). Findings Our mixed methods analyses identified some improvements in educational opportunities for students in classrooms where the eReaders were integrated, including increased access to educational materials, enhanced student engagement and increases in measured academic performance. Our analysis also highlighted multiple barriers to effective eReader integration, such as inconsistent access to electricity, unintended device sharing and difficulties leveraging eReaders to transform instructional practices that could inform efforts to further improve the integration of eReaders in similar low-resource educational settings. Improvements in educational opportunity and outcomes While the number of tablets afforded by Worldreader grant was insufficient to maintain the intended one-to-one student-to-device ratio in all classrooms, seven of the eight interviewed teachers at tablet schools emphasized that the devices increased student access to textbooks. Prior to tablet adoption, as many as eight students shared a textbook. In other instances, only the teacher had access to course material, which he or she used to copy all exercises onto the whiteboard for students to copy into their exercise books. One teacher stated, “In a class environment with no books, the tablets help each pupil to work at their own pace because they each have their own tablet—they can use them anytime. They don’t have to share with anybody.” This represents an augmentation of learning tasks in the SAMR framework (Puentedura, 2013). Students in Kenya who participated in end-of-the-school year focus groups also gave positive feedback on the implementation of tablets in their classrooms and unanimously expressed a preference for tablets over standard textbooks. They highlighted aspects of the tablets such as their ability to efficiently find definitions of unknown words and to access interesting and varied books, and the fact that the tablets did not have missing pages like their textbooks. The students’ perspective likewise illuminates how the tablets augmented learning opportunities, providing functional improvements over the textbooks they replaced. We accordingly observed high levels of digital citizenship or the extent to which students used the eReaders as intended by the instructor. Comments from teachers suggested that the observed behavior reflected students’ respect and appreciation for the opportunity to use tablets and classroom cultural norms regarding teacher authority. Teachers also noted in interviews that the tablets had improved student engagement. Teachers’ evidence for this included decreased student absenteeism and drop-out rates, as well as an observed shift in students’ attitudes toward learning. With respect to students’ physical attendance, one teacher stated that since they received tablets, students were rarely absent. Another provided specific numbers, saying that, “In the past, we had two to three (drop-outs) per term, but this time, they have not (dropped out).” Yet another teacher mentioned that at least three students transferred to the tablet schools from other schools. Teachers attributed these changes to a shift in students’ mindset associated with the opportunity to use tablets. As one teacher explained, “Now pupils like school. Being in school leads to getting something out of that school.” Teachers described students as working more without being told, even without the teacher present in the room, as well as students coming in as early as 6:30 in © 2019 British Educational Research Association 508    British Journal of Educational Technology Vol 51 No 2 2020 the morning to read storybooks on the tablets. Furthermore, they suggested that tablets increased motivation among students in other classes, who attempted to compete with the students with tablet access. One teacher also mentioned that the tablets improved teacher–student relationships by increasing opportunities to communicate with one another, a change potentially leveraged by advances in technological and pedagogical knowledge (TPK) through tablet integration. This was evident in classrooms where teachers could call on more individual students to read or engage in questions in class, since they had access to the text via the tablets. These teachers described tablet use not only changing the context of learning in their schools by redefining students’ orientation to school—potentially because of the status associated with learning via technology in these settings—but also as augmenting prior classroom practices by facilitating communication and improving access to course materials. Here we share excerpts from two classroom observation vignettes that illustrate these enhanced interactions: The teacher called on more than 20 students to read from the eReader. He paused them if they were having trouble pronouncing a word or reading punctuation correctly and would either correct the student or ask the class to correct the student. The teacher would then have that student continue (sending message that it was okay to struggle). The teacher paused to adjust the font size for a student having difficulty. When asking comprehension questions, the teacher holds students accountable to providing text-evidence by asking students to provide the page number and paragraph for where they found their answer [in the eReader]. 42 e-Readers were charged, so most students had one they could refer to, giving them the opportunity to use it for their individual needs (font size, looking up vocabulary words they did not understand, etc.). The teacher has a very good rapport with students. There are many opportunities for participation and quality learning/critical thinking. The teacher takes extra time at the beginning of class to make sure students are on the correct page and that everyone has access to an eReader at their bench (even if they must share with 4 other students). The teacher uses every opportunity he can to have students engaging with the eReader by asking them to read the vocabulary words, read the practice sentences, come up with their own sentences, and work independently, so that students are talking almost the same amount of time as the teacher. The increased access to educational resources and improved student engagement identified in classroom observations and noted by teachers were consistent with greater improvements in oral reading fluency and reading comprehension (in Kiswahili and English) observed among eReader users compared to students in classrooms without eReaders. Our (value-added) regression analysis of the student performance from the baseline to end line assessments shows that students in classrooms with eReaders consistently realized larger increases in academic performance; however, only about one-third of the differences were statistically significant. Table 2 summarizes the regression results, presenting the coefficient estimates for the treatment (eReader) indicator for each of the academic performance measures for the two model specifications (Equations 1 and 2)— separately with pretreatment controls only (for baseline academic performance) and including all controls shown in Table 1—while also adjusting for student clustering in classrooms. The estimated improvements in oral reading fluency and comprehension are larger (and more often statistically significant) for English reading skills. Controlling for student characteristics also increases the magnitude of the estimated differences. These findings are consistent with student comments in focus groups, who self-reported improved grades that they attributed to the tablets, while others cited higher rankings on national exams. Teachers also reported better academic performance among students after receiving eReaders, citing improved and faster reading ability as well as higher achievement in writing, math and science. © 2019 British Educational Research Association 222 198 222 222 222 222 222 198 222 222 222 222 Changes in academic performance from baseline to end line (1) Δ in Kiswahili correct words per minute Δ in English correct words per minute Δ in Kiswahili comprehension Δ in English comprehension Δ in Kiswahili comprehension (incorrect) Δ in English comprehension (incorrect) Predicting end line academic performance (2) Kiswahili correct words per minute English correct words per minute Kiswahili comprehension English comprehension Kiswahili comprehension (incorrect) English comprehension (incorrect) 1.923 7.640 0.260 0.763 −0.423 −0.834 16.539 23.206 0.918 0.883 −0.807 −0.725 Coefficient Note Estimated effects (coefficients) in boldface are statistically significant at α = .05. N Dependent variable 2.265 4.026 0.202 0.187 0.340 0.213 8.502 7.913 0.542 0.504 0.594 0.494 Robust std. error Pretest control only 218 195 218 218 218 218 218 197 218 218 218 218 N 2.636 6.309 0.069 1.054 0.074 −0.878 19.027 25.275 0.726 0.401 −0.656 −0.507 Coefficient 2.946 6.481 0.228 0.322 0.208 0.328 6.256 6.236 0.435 0.420 0.504 0.510 Robust std. error All controls Table 2: Estimated changes in student’s academic performance associated with eReader use (value-added regression analysis results) Effective tablet integration in rural Kenya    509 © 2019 British Educational Research Association 510    British Journal of Educational Technology Vol 51 No 2 2020 Challenges to successful technology integration Despite promising shifts in students’ educational experiences in classrooms with eReaders, several barriers to effective integration limited the extent to which the full potential of eReader use was realized. One of the most pressing concerns raised by teachers was limited access to electricity and related challenges keeping tablets charged. Some, but not all schools, reported access to a generator. Teachers from other schools traveled long distances to charge the tablets at one of the other schools or charged the tablets at their personal residences. While the Kenyan government continues to support rural electrification efforts, further investments in basic infrastructure and the equitable distribution of tablets across all schools will be needed to reduce between-school disparities in tablet access. In addition to charging issues constraining the number of tablets available on a given day, we only observed a one-to-one student to the tablet ratio in 32 percent of the classroom sessions. This limited the ability of students to take full advantage of features that facilitated personalized learning, such as adjusting the font size to improve the readability, working at one’s own pace and taking the tablet home. At the same time, we observed that the device sharing could facilitate peer-to-peer learning and collaboration, indicating that a one-to-one ratio was not a necessary condition for learning with the tablets. In fact, Haßler, Major, and Hennessy (2016) suggest that with the high relative advantage tablets provide many low-resource settings, targeting a oneto-one student to device ratio may not be the best use of limited resources. Instead, the same funds may be better used to enhance the professional development for teachers on device use and integration (Haßler et al., 2016). Indeed, it would have been advantageous to offer more professional training to the rural Kenya teachers on how to leverage tablets for multiple learners working on a single device. The increased use of peer-to-peer learning and collaboration marked one of the most notable deviations from teacher-directed, lecture-based instruction observed in most classrooms. The resulting opportunities for the student agency represent a partial redesign of instructional processes in a manner that began to transform relational dynamics and academic expectations in the classroom. Beyond the opportunities for peer collaboration facilitated through device sharing, the introduction of eReaders was not combined with concerted efforts to assist teachers in transforming instructional practices to be more student-centered. Regardless of the eReader access, most lessons still consisted of teachers copying notes onto the board, teachers lecturing about the notes and students copying the notes or practice questions into their exercise books. It was relatively rare to observe teachers engaging with students (or interacting with the tablets) in a manner that invited student dialogue. Reflecting cultural norms, teachers’ seldom asked students to demonstrate their understanding of the skills being taught until the very end of the lesson (on their homework, checked by teachers after the class). The general lack of (SAMR) transformative practice using the tablets highlights the importance of pedagogical as well as technological knowledge, per TPK intersection (Mishra & Koehler, 2006). This reflects not only a human (instructional) capacity limit... Purchase answer to see full attachment

We offer the bestcustom writing paper services. We have done this question before, we can also do it for you.

Why Choose Us

  • 100% non-plagiarized Papers
  • 24/7 /365 Service Available
  • Affordable Prices
  • Any Paper, Urgency, and Subject
  • Will complete your papers in 6 hours
  • On-time Delivery
  • Money-back and Privacy guarantees
  • Unlimited Amendments upon request
  • Satisfaction guarantee

How it Works

  • Click on the “Place Order” tab at the top menu or “Order Now” icon at the bottom and a new page will appear with an order form to be filled.
  • Fill in your paper’s requirements in the "PAPER DETAILS" section.
  • Fill in your paper’s academic level, deadline, and the required number of pages from the drop-down menus.
  • Click “CREATE ACCOUNT & SIGN IN” to enter your registration details and get an account with us for record-keeping and then, click on “PROCEED TO CHECKOUT” at the bottom of the page.
  • From there, the payment sections will show, follow the guided payment process and your order will be available for our writing team to work on it.