4.1 Participants and Student Organisation
The participating school was a multicultural contributing primary (K-6), located in a middle to low socioeconomic area of a small regional town in New Zealand’s north island. The research class comprised 28 primary (6-7 year old) students who, on average, had been at school for between 1.5 and 2 years. The class comprised 14 boys and 14 girls who were, according to their teachers (Sonia and Lisa, pseudonyms used), of “generally average achievement in reading and writing, according to the PM Benchmark Literacy Assessment” (Sonia, personal communication). The students were familiar with using iPads for other learning, including writing stories using ‘Storybook Creator’ and practising basic addition and subtraction with apps such as ‘Rocket Maths’. They had also used teacher-selected and facilitated sections of the ‘National Geographic Kids’ website to learn about the habitats of native New Zealand animals during a Living World Science unit. However, this was the first time they been taught about and had used online information to create an information artefact.
Sonia was an experienced teacher of 23 years and Assistant Principal, while Lisa was in her 8th year of teaching, 6 of which had been in the junior school. Following Sonia’s recommendation, the students were organised into social pairs for the unit. She suggested doing this would maximise the prospect of quality conversations occurring and increase on task time that would yield more detailed information, as, according to her “they work better with someone they know and are comfortable with” (Sonia, personal communication). iPad display and audio data were collected from all 14 pairs, which, apart from one absence (data were still collected from the remaining student), remained stable across the 6 lessons. The social grouping arrangement was considered acceptable, as the purpose of this study was not to compare students, nor determine any relationship that might exist between, for example, existing reading capabilities and their ability to successfully complete this task. Instead, this study sought evidence to determine whether students of this age are capable of understanding and operationalising basic online research and information evaluation and communication capabilities, when taught using a project-based curriculum. Its principal purpose was to investigate the efficacy of this approach for early years classrooms.
4.2 The Teaching Unit
The students were tasked with creating an information artefact (digital representation) for a specified audience (their ‘buddy’ class), reporting at least 3 facts about Matariki (the Maori New Year) resulting from their online research.
The learning unit was described by Sonia as:
… an integrated language and digital technology unit, where the children need to use the web to research and then design a Pic Collage[1] poster or Popplet[2] map with 3 facts about Matariki, to share with their buddy class using the IWB (interactive whiteboard)… (Sonia, unit plan).
Learning outcomes aligned with New Zealand’s national curriculum objectives in Digital Technologies (DT) and English (Eng.) at Level 1 (Ministry of Education, 2007). These were:
- Designing and developing digital outcomes (DT);
- Design and visual communication (DT);
- Acquire and begin to use sources of information, processes and strategies to identify, form and express ideas (Eng.).
To help guide planning, after discussion with the teachers, Durrant and Green’s (2000) original Technoliteracy dimensions were revised to acknowledge the close relationship existing between Cultural and Critical dimension capabilities. To accommodate this the dimensions were combined, reflecting one of the study’s main objectives which was to learn about the young students’ ability to critically review – even at an emerging level, the information they were accessing, and display awareness that it represented particular people’s perspectives on events, which may or may not be accurate. The combined dimension also accommodated the unit’s learning objective to demonstrate basic understanding of appropriate design, layout and use of colour in producing an information artefact to be communicated to a specific audience. While these objectives may appear broad and ambitious for a single teaching unit, it is important to remember it was not the intention of this study to attribute a specific measurement to how well students could execute capabilities within the dimensions, or assess any development of these during the teaching unit. Given their young age and some literature suggesting such capabilities are ‘developmentally inappropriate’ for young students to learn (see Share, 2010), its principal purpose was to investigate the validity of these claims, given increased device access and use by much younger children.
4.3 The Lessons
The lessons were planned using the revised and expanded dimensions and capabilities from Durrant and Green’s (2000) model (see Figure 3). They focused on:
• basic web search skills including using keywords and strings. Examples were recorded on the class whiteboard;
• synonyms, as equivalent keywords;
• how to use search results page descriptions to judge the potential relevance of information;
• how to use the assistive functions and cognitive tools of the iPad to access text they couldn’t read or to check their own understanding, and to improve the accuracy of writing (e.g., text-to-speech, auto-correct, word picker, spell checker, predictive text);
• how to copy and paste text or images between apps and use the ‘app switcher’ function;
• how to determine what is likely to be ‘true’ by locating similar information on more than one website (evaluating, verifying);
• the internet is an ‘open’ environment representing people’s views and perspectives, and that these may not always be ‘true’;
• desired qualities of the digital artefacts that make them ‘fit for purpose’ (i.e., presenting to their ‘buddy’ class). These included layout and visibility attributes for presenting to an audience at a distance, using an IWB.
Of note is that in planning and teaching, the above capabilities were not introduced sequentially- that is, on a ‘lesson by lesson’ basis. That was because the unit was designed using a project-based approach, with specific capabilities and understandings being introduced at relevant times during the artefact development process, as opposed to being time-bound within a particular lesson. The ‘just in time’ instructional approach was intentional, to maximise opportunities for the students to immediately practise and apply the taught capabilities thereby enhancing their relevance, and potentially, their understanding.
Each 40-45 minute lesson followed a similar structure. A 15-18 minute (approx.) whole class introduction was teacher led and comprised various combinations of: revision of previous lessons; direct instruction; modelling and demonstrating task-relevant knowledge and skills; creating learning scaffolds (e.g., recording keywords/strings or question stems on the whiteboard); discussing and demonstrating strategies for checking the accuracy of web-sourced information - and why this is important, and building general understanding of the web as an open information environment. In the remaining lesson time, the pairs completed their online research and artefact authoring using the Pages app as an editing tool, before finalising their 3 facts into either PicCollage or Popplet. As the students worked, both teachers circulated around the pairs providing guidance and direction, as required.
4.4 Data Collection and Coding
Data were collected using the native display and audio capture app available in iOS 14, although a tweak was made to its operation so that it continued recording even if the students closed the iPad’s cover (which occurred frequently). Recordingsfor each pairwere collected across the 6 lessons, and were organised in ‘bundles’ of between 2hrs 16 mins and 2 hrs 51 minutes of runtime duration. These data included teacher-student (advice, guidance etc.) and student-student on and off-task actions and dialogue, but excluded introductions and any plenary discussions. Data coding involved 2 coders, and followed conventional inductive methods. First, a sample comprising 8 hours 24 minutes of recorded data was manually reviewed by the author and a postgraduate research assistant (RA) working together, to identify occurrences considered to align with the capabilities outlined in the revised analytical model. The sample was randomly selected, although care was taken to include some data from each pair. Although time consuming, this process was necessary to enhance the accuracy of decisions when coding the full dataset. Furthermore, it was decided that occurrences would be coded as percentages of recorded runtime, rather than as single event counts. This was considered more appropriate, as V-Note timeline analysis indicated occurrences often spanned several seconds or sometimes minutes rather than being ‘one off’ events, as students discussed, negotiated and trialled different strategies (see Figure 4).
During the initial analysis, the author and RA discussed categorisations and the ‘goodness of fit’ of data with the revised model dimensions and capabilities, using the descriptions recorded in Table 1. While general compatibility was noted, analysis suggested demarcation between dimension capabilities was not as clear cut as represented in the model, but that activities existed between them where capabilities interacted in processes of meaning making, meaning sharing and communication, and assessing meaning quality. Where this was evident, rather than trying to apportion ‘weightings’ the overlaps were tagged and recorded on the analysis timelines, along with notes describing the interactions and the influence they had on the students’ work processes. Following the initial analysis and refinement of data classification descriptions the RA timeline coded the full dataset, including the previously coded sample. In total, 37 hours, 27 minutes of recorded data were coded. Although exceptionally time-consuming, coding the full data set was essential to ensure accurate coding decisions, given the relatively small number of participants. When completed, a randomly selected sample (20%) of coded data was re-coded by the author, and interrater agreement (kappa) calculations performed. To make this process manageable, data were reviewed against the larger dimensions of the model, rather than using individual capabilities. Predictably, while substantial agreement was evident in data coded in the Operational dimension, more modest (although still acceptable) results were apparent in the Knowledge-building and Cultural-critical dimensions. This could be attributed to the more easily identifiable data where technical/operational capabilities were being used, as opposed to subjective interpretation of data aligned with the other two dimensions. Table 2 contains a summary of interrater agreement results.
[1] See: https://piccollage.com/
[2] See: https://www.popplet.com/