1. Start with your valued learning outcomes, and the holistic achievement of your students. The data you collect, your analysis and ultimately the strategies you put in place should all flow from your valued learning outcomes. Track where students are at as well as the progress they are making in relation to valued outcomes.

2. Make data and data conversations public. Data analysis should not be the responsibility of a single person in a school as no one can individually identify all of the patterns in the data. Bring groups of staff together to develop a shared understanding of data. Encourage people to ask questions, to challenge assumptions, and to check their preconceptions and blindspots.

3. Always ask “What’s your evidence?” We all approach data with particular preconceptions and beliefs but it’s important not to default to assumptions, preconceptions or hearsay. Make sure that you dig deeper and find out what’s actually happening in reality rather than assuming what’s going on.

4. Don’t rely on a single source of data, or over privilege certain types of data (especially standardised tests, which are only ever part of the puzzle). Relying on a single data source does not capture the richness of all that you want your students to achieve. Furthermore, no single test can ever tell you everything that you need to know about a student’s performance in a particular subject area. For instance, a PAT reading comprehension test, while telling you something about your students’ reading ability, does not provide data on attitudes, their ability to engage with a wider range of texts, or how they read in situations other than a time-bound test.

5. Don’t default to averages. Attributing too much to means and medians (measures of central tendency), without also looking at variability in the data, masks the complexity of what is going on in students’ learning. It’s important to disaggregate data and to drill down into the data by looking at different levels or groups, for example, by year level, individual class, ethnicity and gender.

6. Take the time to properly define and understand the problem. We often want to jump quickly from identifying an issue of student achievement to implementing solutions. However, it’s important to engage with all interested stakeholders, including school leaders, teachers, students and whānau, to talk about all the factors that could be contributing to the issue. Then, systematically test each factor by collecting further data and evidence.

7. There are no silver bullets. No programme or approach is perfect all the time. When selecting a “solution” make sure that there is a good fit between your valued learning outcomes, the patterns of students’ achievement outcomes and the possible solution. There is no one-size fits all programme.

8. Improvement takes time. While there are often some fine tweaks you can make that will lead to “quick wins”, it can often take 1 to 2 years for significant shifts to occur. It is important to monitor student achievement throughout this period. This should be done formally at the beginning and end of each year as well as informally through small inquiry points throughout the year. These inquiry points could include teachers assessing and reviewing student work, collecting student voice through conversations and surveys and engaging in critical reflection.