'Investigating Stereotype Threat in the UK Junior Mathematical Challenge' (AsPredicted #21450)
Author(s) Matthew Inglis (Loughborough University) - m.j.inglis@lboro.ac.uk Steven O'Hagan (UKMT) - S.OHagan@leeds.ac.uk
Pre-registered on 03/28/2019 08:34 AM (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? Do women perform better on mathematics tests if they are asked about their gender after completing the test rather than before? More precisely, we aim to closely replicate the study conducted by Stricker & Ward (2004) and discussed by Danaher & Crandall (2008) and Stricker & Ward (2008). While Stricker & Ward also focused on ethnicity, we will only consider gender.
3) Describe the key dependent variable(s) specifying how they will be measured. Performance on the UK Junior Mathematical Challenge scored in the standard way.
4) How many and which conditions will participants be assigned to? Two conditions: (i) participants will be asked to state their name and gender in advance of taking the test, and their school name and school year after taking the test; (ii) participants will be asked these questions in the reverse order (i.e. to state their school name and school year before taking the test and their name and gender after taking the test). Participants will be assigned to conditions randomly.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. We will follow Stricker & Ward’s approach by conducting a 2 (gender) 2 (condition) Analysis of Variance (ANOVA), followed up by Scheffe post-hoc tests. If the manipulation influences female participants’ performance then we would expect a significant gender by condition interaction, with girls performing worse in condition (i) than in condition (ii).
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. See box 7.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. As of 28th March 2019, a total of 6 schools have agreed to participate (approx. 2180 students). They will be sent detailed instructions for how to administer the UK Junior Mathematical Challenge, which takes place on 30th April 2019. After the administration we will ask schools to verify that they followed the instructions to the letter. If they confirm that they did, then all their students who consented for their data to be analysed will be included in the final analysis. Students from schools which are unable to confirm that they followed instructions will be excluded from the analysis.
All data will be collected as part of the annual UK Junior Mathematical Challenge on 30th April 2019.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) Secondary Analyses:
(i) Alongside the traditional frequentist analysis we will also conduct a gender condition Bayesian ANOVA, with small and medium r scales (which we will interpret as 0.2 and 0.5).
(ii) We will assess whether a multilevel model approach would be appropriate (given that participants are taking the tests in examination conditions we suspect this will not be necessary). We will do this by running a linear model where scores are predicted by gender, condition and the gendercondition interaction effect, and assessing whether allowing intercepts and slopes to vary between schools improves the fit of the model. Any single-sex schools who participate will be excluded from this analysis.