This research used a scoping review, which is methodologically similar to a systematic review, to provide a rigorous knowledge synthesis of current evidence.25-26 For the purpose of this study, the scoping review framework used was described by Arksey and O’Malley27 as a five-step process with an optional sixth step. These steps include: (1) identifying the research question, as the starting point and as the launching point to guide the search strategy; (2) identifying relevant studies, which involved the development of a comprehensive search strategy to ensure accurate and complete results; (3) selecting studies, which involved developing a-priori inclusion and exclusion criteria that were revised throughout the review, as familiarity with the evidence increased; (4) charting the data, which involved charting and sorting key material from the results into themes and trends; (5) collating, summarizing, and reporting the results, which involved presenting the results as a visual and/or narrative; and finally, (6) consulting with relevant stakeholders, which is contingent upon time and resource considerations. For the purposes of this research, the sixth step was not performed.
Review Protocol, Team, and Management
To ensure transparency, rigour, reproducibility, and consistency, protocols were developed prior to the start of the research, for the inclusion criteria, search protocols, and data characterization utility form. This helped to ensure an unbiased approach to the search protocol and to enhance rigor.27 These are available upon request. The scoping review was conducted by a team of individuals with multi-disciplinary capability in nursing, knowledge synthesis methodologies, and ICTs. The primary reviewers included the primary and secondary authors, as well as one research assistant. In addition, a University librarian was consulted throughout the search term selection process to ensure completeness and accuracy of search terms contributing to a comprehensive and complete search strategy.
Any and all potentially relevant citations identified throughout all stages were imported into EndNoteTM, a reference management software, where duplicates were removed by the program and then double checked, with manual removal by the primary author; the list of citations was then imported into a web-based electronic systematic review management platform, DistillerSRTM. The screening for article relevance, up to the data extraction stages, were conducted within this software. Two reviewers (the primary author and research assistant) were used throughout the selection and analysis process to ensure consistency, adherence to the inclusion/exclusion criteria, relevance to the research question, as well as the categorization of data into themes and trends. As part of this process, all articles were screened by the primary author and research assistant. Any discrepancies were brought forward to the secondary author who made an independent decision whether to include or exclude.
Review Question & Scope
The authors used a scoping review methodology to address the following question: What is the global impact of ICTs on women’s empowerment?
Search Strategy
The authors ensured identification of relevant and suitable publications by creating a search strategy protocol prior to retrieving evidence from a variety of sources. As per Arksey and O'Malley27, the following avenues were reviewed as part of the search strategy: searching relevant electronic databases, reviewing reference lists of pertinent articles to identify addition sources, and manually searching key journals.
To ensure the search was comprehensive, the following databases, available via the University of Saskatchewan library, were searched on November 30, 2016 and updated on January 1, 2018: Scopus, Embase, ABI Inform, Soc Index, Sociological Abstracts, Gender Studies, Springer Link, PsychInfo, Science Direct, and Academic Search Complete. The COCHRANE Library was also searched for any relevant trials in the trial registry. Limits placed on the search included: English only, no book reviews, publication dated 2012-2017, and the protocol was pretested in Scopus and Soc Index using select key words including “women” and “empowerment” and “technology.” An illustration of the search term strategy can be found in Table 1.
Table 1: Search Term Strategy
Women Search Terms
|
|
ICT Search Terms
|
|
Empowerment Search Terms
|
*searched with OR
|
“AND”
|
*searched with OR
|
“ AND”
|
*searched with OR
|
Wom?n
|
|
Technolog*
|
|
Empower*
|
Female*
|
|
Information technolog*
|
|
Disempower*
|
Girl*
|
|
"information communications technolog*"
|
|
Barrier*
|
Maternal
|
|
"ICTs"
|
|
Enable*
|
|
|
"social media"
|
|
Self concept
|
|
|
mobile
|
|
Self efficacy
|
|
|
handheld
|
|
Capacit*
|
|
|
telehealth
|
|
Emancipat*
|
|
|
computer
|
|
|
|
|
Smartphone
|
|
|
|
|
Digital
|
|
|
|
|
Internet
|
|
|
|
|
Telecommunication*
|
|
|
|
|
"world wide web"
|
|
|
|
|
Laptop
|
|
|
|
|
ICT4D
|
|
|
|
|
“web-based”
|
|
|
|
|
Iphone
|
|
|
|
|
Ipad
|
|
|
Legend: * at end of word = truncation, any number of letters (e.g. capacit* will find
capacity or capacities); ? at end of beginning of word is used to represent one or more other
characters in a search term (e.g. wom?n will find women or woman).
Limits included: 2012-2017, English language, no books/book reviews
Search terms were drawn from the research question, as well as from lengthy discussions with the university librarian and expanded upon based on a cursory search of two databases. To determine the range and breadth of key terms, an initial limited search of two databases was conducted yielding several papers. These databases were determined in consultation with the university librarian and included Scopus and Gender Studies. These papers were then analyzed for similar keywords, definitions, analogies, and index terms that were relevant synonyms to the initial search words.28-29 These additional terms were added to a master list that informed the final search strategy. Specifically, for the term empowerment, keywords were chosen that could provide results that included a lack of empowerment as well, thus the inclusion of “barrier” and “disempower”. The other search terms came directly from key articles and databases and were demonstrated to be the most common variations on the term “empower”. An additional term that was used interchangeably with “empower” was “agency”, however, as this term is used more frequently in conjunction with organizations and not empowerment, it was removed from the search term list.
The ability of the electronic database search to identify all relevant primary research was verified by hand searching the reference lists of eight key peer reviewed articles and nine key electronic journals that were flagged through the initial test search as well as the main search. The journals were chosen based on their relevance to the research question as well as their scholarly nature. The initial three identified journals were: Community Informatics, Gender and Development, and Journal of Women in Culture and Society. Subsequent journals were identified and selected for a hand-search once the initial search was completed. These were: Gender, Technology & Development, Computers in Human Behaviour, American Journal of Health Behaviour, American Journal of Public Health, and Women’s Health Issues. These journals were then reviewed for additional articles potentially not identified via the database search; this included entering the general search into journal databases.
Additional grey literature was identified by hand-searching the websites of the Association for Computing Machinery Digital Library Journals and Conference Proceedings, the UN Women, Status of Women Canada, the United Nations Development Programme, the International Center for the Research of Women, the Girls Action Foundation, the Information and Communications Technology Council, the International Telecommunication Union, and the International Development Research Center for primary research reports, guidelines, situation reports, and referenced publications that were not already included.
Study Selection: Relevance screening and inclusion criteria
The focus of the study selection was locating published and unpublished academic articles, which may have any type of study design, including qualitative, quantitative, or mixed methods. The initial pool of results included a total of 4481 citations. An initial set of inclusion and exclusion criteria were developed a-priori to screen abstracts and titles of citations and were refined during each review of the pool of articles. Research articles were initially considered relevant if they included women’s empowerment and/or information and communication technology concepts in the title or abstract of the publication. Synonyms for these concepts were created in consultation with the librarian to ensure a robust search strategy for maximum location and inclusion of studies. Given the evolving nature of ICTs and their role in interventions, the authors wanted the articles to reflect a recent knowledge base, therefore the timeframe of 2012-2016 was chosen, which was later expanded to December 31, 2017 as the review progressed. The results were also filtered to include English only content.
First Screen: Inclusion criteria
The inclusion criteria created for the first level of study selection were driven by the research question topics, specifically, women, empowerment, and ICTs. According to the JBI (2015), the inclusion criteria should be based on three themes, also known by the acronym of PCC: (a) participant description, (b) concept, which is likened to the phenomena of interest, and (c) context. The inclusion criteria used in the first level of selection were country of publication, date of publication (2012-2017), and the use of both of the following concepts in the title or abstract of the publication: women’s empowerment and/or information and communication technology. At this stage, the primary author looked for the presence of the key words in the title and/or abstract. The use of these keywords as inclusion criteria were designed to be intentionally broad to provide a sense of what publications linked the two concepts (i.e., women's empowerment and ICTs).
First Screen: Study selection
On first review, the initial pool of articles was subjected to a staged process to ensure studies were selected that were relevant to the research question and met the inclusion criteria. Articles were first excluded based on duplication within the initial search results. This exclusion was conducted using the search tools feature within the electronic database, but also within the reference management program Endnote™ and then manually by the primary author. The inclusion criteria were applied to the title and abstract of the publication. Any title or abstract that did not meet the inclusion criteria was removed from further review and consideration. All articles excluded by the criteria were sent to the research assistant who confirmed the exclusion. Any disagreements or contradictions between the primary author and the research assistant were thoroughly discussed, with both parties having to agree to the inclusion before the publication would be added back into the pool of articles to move forward. Additionally, if an article could not be excluded based solely on the title or the abstract, the full article was reviewed for relevance to the research question and inclusion criteria. These latter two points did not prove to be an issue as there were no disagreements.
Second Screen
The remaining pool of articles was then reviewed a second time by applying a second level of inclusion criteria to the title as well as the abstract. It is common and encouraged as part of the scoping review process to generate increased cumulative familiarity with how concepts are presented within the evidence. This, in turn, informed the decisions that were made regarding the inclusion or exclusion criteria of the subsequent stage. Much of the articles after the first level of elimination included technology as a passive aspect of the study and not one that women actively participated in. It was important for the authors that the technological aspect of each study be an intervention that women could engage in towards building self-efficacy and capacity. This informs current gaps within the evidence that speak to how women are using technologies to support their empowerment. As such, this set of inclusion criteria focused on technology as an intervention in the study and women as active participants in the study versus just the word “women” found throughout the first set of criteria.
Final Screen
For the final review of the full text articles, based on the content and findings in the scoping review process, an additional criterion was included. The authors wanted to explore how the social determinants of health informed and supported the concepts of women, empowerment, and ICTs. At this stage, it was noted which social determinants of health, if any, were present in each article. The list of social determinants as per the Government of Canada (2019) was utilized as a reference for this portion of the process. The remaining 59 articles all had social determinants of health present. After another review using the above inclusion/exclusion criteria, 14 more articles were removed from consideration. Rather than focus on a range of these determinants, the authors decided to include all 45 articles and to then review the implications of this finding in the analysis (Figure 1).
Figure 1: PRISMA Flow for Screening Process
Re-run Searches
Due to the gap in the timeline as to when the review began, the authors included re-run searches for each database up to January 1, 2018. A total of 573 articles were found in all 10 of the main electronic databases. Through the first cut and second cut, using the same above inclusion and exclusion criteria, all but six new articles were eliminated. The final total of articles included within the analysis was 51.
Study Characteristics, Extraction, and Charting
The final step required of Arksey and O’Malley’s27 scoping review framework was to collate and summarize the results for presentation and discussion. Each selected article was summarized in a customized data characterization utility form to guide data extraction. The goal of this step was to determine and chart factors to be extracted from each article to help answer the research question.26-27, 30 The charting of data was an iterative and exploratory process in which the data charts were continuously updated to ensure completeness and accuracy.26, 30 Data extracted from the charts included year of publication, country of study, implications for policy and practice, types of ICT interventions, demographics, empowerment (definition, as a design consideration, and measures), and social determinants of health (presence and description within in the study). All data was then analyzed using thematic analysis and the main ideas refined over several iterations. The data was then mapped using tabular and visual presentations of the main conceptual categories followed by a narrative summary describing how the results related to the research question and objectives.