METHODOLOGY
3.1. Research Framework Overview
To identify the factors that need to be considered when designing e-learning environments to engage elderly learners, I employed a research framework that exploited a qualitative strategy. This strategy was predominantly inductive in approach. Within this framework, observations, oral reports and interview techniques were used for data collection purposes. Due to the comparative and iterative systems running through the study, multiple case studies were employed. “A single case study would concentrate on that case alone. More complex, multiple case studies might involve several such individual cases.” (Robson, 2002; p.180). A helpful analogy put forward by Yin (2003) suggests that conducting multiple case studies is akin to carrying out multiple experiments. These might be endeavours at reproduction of a first experiment; or they could feed into and add to the initial experiment, possibly bringing the exploration into an area dictated by the first study; or perhaps endeavour to complement the first study by investigating a different aspect. The initial case study, still supplies “evidence” which sustains a theoretical perspective about what is occurring, maybe in relation to mechanisms and the contexts in which they operate (Robson, 2002, p183).
Data extracted from the recordings and transcripts of each of the individual case studies were compared with the data extracted from the other case studies. This inter-analysis of cases allowed codes and patterns to be identified and assisted in the identification of the important factors that need to be considered when designing e-learning environments to engage the elderly learner. Moreover, the identification of themes at a ‘latent’ level (see Boyatzis, 1998; Braun & Clarke 2006 ) enabled an examination of the “underlying ideas, assumptions and conceptualizations that were theorized as shaping or informing the semantic content of the data.“ (Braun & Clarke 2006 p.84).3.2. The Test & Participants
Data drawn from four learners who participated in four diverse Web-based learning activities contributed to this study. Studies by Virzi (1992), Nielsen and Landauer (1993), and Lewis (1994) illustrate that only a small sample size is necessary to run effective usability tests. They advise that the first three to five testers expose most usability problems. The ages of the learners in the present study ranged from 63 to 87 years old. In an effort to find participants, I contacted the manager of the Terenure Enterprise Centre in Dublin in October 2010. The manager of this Centre subsequently requested that the Active Senior IT Society group include my request for participants in its newsletter. By way of this advertisement, I received a number of responses via telephone and email. From these contacts three of the participants were selected. I personally approached the oldest participant and I invited him to participate based on age criteria and fitness levels.
The tests were carried out over a three month period. The participants were requested to complete a number of exercises within four different learning activities. A detailed description of these activities is provided in chapter four. The learners’ interactions with these activities were observed and the participant learners provided oral reports on their thoughts and feelings as they interacted with the activities. Recordings of screen interactions and voice recordings were taken. The learner who participated in the first test retested in the final test three months later i.e. the fifth test. The final test enabled an examination of the impact the redesigns had on a learner’s engagement levels. Moreover, it acted in a confirmatory capacity in relation to themes and factors identified. The final test took place in the same home environment as did the first test. After each participant had completed all the activities, they then participated in a semi-structured interview. The duration of the individual test sessions ranged from one and half hours to two hours.
3.3 Research Ethics
Before I outline the techniques employed in gathering data, I feel it is necessary to highlight the importance of research ethics. Ethical issues were of principal concern throughout the study. The participants have confidentiality rights and anonymity rights that must be provided for (see Laws, 2003). Further, participants and respondents should not be hurt in any way by the research either during participation or subsequent to participation (see Gray, 2004). In the present study, confidentiality has been maintained and the participants will remain anonymous. In addition, permission to record the participants’ interactions with the activities and their oral reports was given by each of the participants. Furthermore, these recordings did not permit the identification of the learners and they have been employed for the sole purposes of this study. The recordings have since been disposed and therefore, cannot be employed in further studies.
Observations
Oral Self Reporting
Post Test Semi-structured Interviews
3.4.1. Observations - Screen Interaction Recordings
Observations and screen interaction recordings allowed a close examination of how the different learners interacted with the learning activities. Behavioural responses and physiological responses such as posture, facial expressions and general body language provided clues in relation to the important factors that need to be considered when designing e-learning environments to engage elderly learners. They also provided clues in regard to the level of engagement learners were experiencing while interacting with the learning activities. However, since the observations were obtrusive the participants knew they were being observed, and this may have effected how they interacted with the activities. People may have a tendency to alter their behaviour when they know they are being observed – this is known as the observer effect (see Thompson, 2009). A method to avoid the observer effect is to use unobtrusive procedures (see Webb et al, 2006) whereby, the participants are unaware they are being observed. However, due to the limits of this study, this was not possible. To counteract the probability of the observer effect – to some extent at least - self-reporting through voice recordings and interviews were also employed and as result acted as check on any observer biases that may have interfered with a true depiction of events. Another threat to the accuracy of conclusions arrived at is the fact that observational research draws on the subjective judgments of the observer in reaching conclusions. Therefore, conclusions drawn by one observer might not be the same as the conclusions drawn by another observer (O’Grady, 2001).3.4.2. Oral Self-Reports
Another useful method employed in this study was via real time oral self-reports. This method was influenced by the Experience Sampling Method (ESM). ESM was developed by Csikszentmihalyi & Larson (1987). It is a quasi-naturalistic method that involves the subjects of a study to report, at random times throughout a period, on their experience in terms of type of experience and the quality of experience. In the present study, ‘thinking out loud’ enabled real time oral reports by the learner as she or he proceeded through an activity. Recordings of these reports allowed examination of the learners’ utterances as they interacted with the learning activities.3.4.3. The Semi Structured Interview
Interviews are perhaps the method most utilized by a researcher in a qualitative study. An important characteristic of the interview is its capacity to concentrate intensely on the respondent (see Ritchie & Lewis, 2010). “The role of the researcher is that of a facilitator to enable the interviewee to talk about their thoughts, feelings, views and experiences.”(see Ritchie & Lewis, 2010 p,146). The interviewing technique employed by this study used open ended questioning, which allowed the respondents enough flexibility to respond in his or her own way and enough interviewer control to restrict respondents from diverging. They also provided a space for clarification and further elaborations and allowed probing by adding to the initial question e.g. with add-ons such as, anything else, why is that and/or what do you mean etcetera. The semi-structured interview permits comments to be explored that go further than the initial question posed (Cresswell, 2003). The weakness of the interview is that interviewees may have tried, at times, to give responses that they felt were more acceptable in the interview, which may have been contributed to by the initiating role of the interviewer in posing questions and requesting clarifications. Therefore, it was importance to evaluate in the analysis stage whether this had occurred.3.5. Analytic Framework
Thematic analysis provided an analytic framework which facilitated the identification of themes (see Boyatizs, 1998; Braun & Clarke, 2006). This method is “effectively independent of theory and epistemology”, and has the capacity to traverse a range of theoretical and epistemological strategies and this makes it a highly flexible and useful approach to analysis (Braun & Clarke, 2006 p.78). In other words, “Thematic analysis is not wedded to any pre-existing theoretical framework and therefore, it can be used with different theoretical frameworks.” (Braun & CLarke, 2006 P. 81). The method was chosen not only because of its independence of pre-existing theory, but also for the fact that it does not restrict the research to generate a wholly built-up grounded-theory analysis (sees Strauss & Corbin, 1998; see Braun & Clarke, 2006). Thematic analysis provided the flexibility which allowed themes and patterns to be evaluated and compared across the dataset. Moreover, analysis at a latent level permitted the examination of themes to go beyond the immediate surface setting (see Boyatzis, 1998), and to explore the data within the context of an elderly persons’ psychosocial status, physical and cognitive capacities as well as within the context of their emotional and psychological needs.
3.5.1. Steps In The Analysis Of Qualitative DataTranscription of key communications from observations, oral reports and interviewsNotation of initial ideasGeneration of initial codesSearch for themes (patterns)Review of themesDefinitions and labelling of themesProduction of report
(see Braun & Clarke, 2006 P. 87 )First and foremost, it should be noted that analysis of the dataset was undertaken for two reasons: the primary reason was to identify, through thematic analysis, the factors pertaining more directly to the question, while the second reason was to develop the Web-based learning activities. In the initial stages of a study, key pieces of the aural and visual data from the screen interactions, oral reports and interviews were transcribed and notes/memos were used to record initial thoughts and ideas of what was going on. To follow, almost everything from the transcriptions was coded. In the search for themes, codes were systematically examined, first within cases and then across the entire dataset. Relationships, patterns, similarities and differences in the data were identified and grouped together accordingly under identified themes. Although the research was primarily inductive in approach, both inductive and deductive reasoning was called upon when re-examining the findings from one case within one interation with the findings from another case within another iteration. The data analysis involved going back and forward over the recordings in attempt to re-examine and compare data with the aim of refining and confirming themes. The data which signified one theme may also have signified another theme. However, the same data analysed across themes had different meanings within the different themes and also had different meanings within the diverse contexts of a particular theme. The final stages of analysis entailed generating clear definitions and labels for each theme. This system also allowed secondary sources to be explored in context of the themes already identified inductively and thereby, add support to the analysis. The process of analysis allowed me, as researcher, to gain a deep understanding of the entire dataset and a sensitized perception of the important factors that need to be considered when designing e-learning environments to engage the elderly. In line with the expectations of thematic analysis a report was produced based on the researcher’s interpretations and analysis of the dataset. This report presents and discusses the findings from the empirical research together with the findings from secondary sources.3.6. Chapter Summary
This chapter outlines and discusses the methodology employed in identifying the factors that are important to be considered when designing e-learning environments to engage elderly learners. The chapter presents and critiques the techniques employed in gathering the research data. An overview of test sessions is provided and the participant learners are introduced while at the same time anonymity and confidentially is sustained (see Gray, 2004; Laws, 2003). The final sections provide details of the analytical framework wherein thematic analysis is discussed and the key roles note-taking, coding, and themes play are detailed. Emphasis at this stage is also on the importance of final report and the interpretations of the researcher.
Copyright @ FH