Master's Dissertation First Full Draft

4.8. Procedure

After Departmental and Faculty review of the research proposal, an application for ethics clearance was submitted the Wits Human Research Ethics Committee (non-medical) – HREC (non-medical). In parallel, a research approval request was submitted to the Office of Knowledge Management and Research, Gauteng Department of Education (GDE). Approval was granted by the GDE (ref. no. D2016 / 178 GA) and ethics clearance was subsequently granted by the HREC (non-medical) – protocol number H15/05/41 to conduct this research. As stipulated in the GDE research approval letter, this letter was forwarded to the relevant district directors via the district co-ordinators of the Secondary School Improvement Programme (SSIP).

A pilot study with a small number of participants was then conducted at a science centre with students participating in an enrichment programme run by the centre. Permission for this was first obtained from the CEO of the centre. Feedback from this pilot study was incorporated and necessary changes were made to the instruments and the experimental procedure.

At all six sites where these experiments were conducted, researchers were allocated a classroom space to make use of. All of these classrooms were largely similar in terms of their size, lighting, types of desks and chairs, environment (i.e. temperature and humidity) and external noise level (low to moderate) – representative of real-world learning conditions.

At sites 1-5 participants were assigned to a device and reading/note-taking condition and given 30 minutes to study the text passage assigned to them. Before beginning their assigned task, participants first completed the demographic questionnaire. Approximately one week later, participants returned, were given 5 minutes to revise the text and/or their notes, followed by a 10-minute test on the text passage. For participants who completed only one device condition, this marked the end of their research participation. Participants completing multiple conditions, immediately after completing their first test, were then assigned to another device condition and the same reading/note-taking condition and given 30 minutes to study a different text. These participants then returned approximately one week later, were given 5 minutes to revise the text/their notes followed by a 10-minute test.

Participants at site 6 followed a nearly identical procedure to the one device only sample. The only difference was that half of site 6 participants were given 2 days between initial studying of text and test, while the other half were given 3 days. This was enforced by logistical limitations – this was the only available time frame which could be allocated by the organisers of the site 6 timetable and compares to the approximately one week between initial reading and test that occurred at the other five sites. At all six sites, each experimental session lasted approximately 50 minutes, both for participants only completing one device condition and for those completing two or more device conditions.

Two of the six sites where experimental data was collected are SSIP participants and are located in two separate Gauteng townships. This programme is structured so that pupils from approximately two to three schools (with this number depending on a range of factors) attend classes at one of those schools on Saturdays and during school holidays. Although primarily targeted at underperforming schools, many schools on this programme have seen dramatic performance improvements in a short space of time and are no longer considered underperforming. The two SSIP schools where experiments were conducted both have matric pass rates of 90%+ and are therefore considered to be fairly high performing institutions.

At both these SSIP sites, research was conducted on Saturday mornings, over a period of two Saturdays at one site and four Saturdays at the other. In both cases preparations were made with the relevant SSIP site manager beforehand. Learners who had free slots in their Saturday morning class timetable were invited to participate in the research. The SSIP programme involves matric learners and participants from these two sites were 18 years or older, allowing participants to give informed consent themselves, without requiring informed consent from their parents or legal guardians. This approach was requested by district officials to avoid researchers having to approach learners during the week. The site managers allocated an empty classroom for use by the researchers. At site 5 participants were assigned both to a device condition and a particular text passage on the first Saturday morning – all participants completed the reading only condition. At site 3, participants had each been provided with a tablet by the GDE and as such learners were allowed to use their own devices. Therefore, on the first Saturday morning at site 3, all participants completed the paper note-taking condition using the same text and the following week, participants all completed the tablet note-taking condition, using a different text, but with all participants using the same text. Due to a number of factors, including an alternate week subject scheduling on learner timetables, relatively high rates of attrition were experienced at this site, necessitating two additional sessions at site 3. Despite four sessions, very few participants successfully completed two conditions, leaving most of the data collected from both these sites as one device condition only.

Sites 2 and 4 are both suburban secondary schools, while site 1 is a private urban school. Research at these schools was conducted during the week, at times allocated by the school management. At all three schools, grade 11 learners were addressed as a single group and invited to participate in the research. Participant information sheets and consent/assent forms were also handed out to learners. The researchers then returned approximately one week later and began conducting the research experiments over a number of weeks. Participants at sites 1 and 4 completed the paper condition first followed by the tablet condition a week later, primarily due to practical limitations. Participants at site 1 all completed the same text passage during the first session and participants all completed the same text (different to the first week’s) during the second session. However, the majority of participants at this site did complete two device conditions, allowing for comparisons to be made within samples. Participants at site 2 completed the same text passage each week, but were ‘quasi-randomly’ assigned to device conditions, leading to a fairly balanced distribution of device conditions and text passages.