This article provides clear and practical guidelines for researchers seeking to use national secondary datasets to conduct evidence-based research. Drawing from our own experiences, we discuss a six-step research process model (Bryan et al., 2010, 2017) to help researchers navigate the use of these datasets. We present examples from the school…

This study demonstrates how to evaluate a university-wide online course designed to support student transition into university by using Propensity Score Matching (PSM) and Doubly Robust Estimation (DRE). Using data from seven academic years, from 2016/17 to 2022/23, with more than 28,000 students, we examine whether enrolment in this optional…

Recently, deep learning has become a pervasive tool in prediction problems for structured and/or unstructured big data in various areas including science and engineering. In particular, deep neural network models (i.e. a basic core model of deep learning) can be viewed as an extension of statistical models by going through the incorporation of…

This article presents methods for using extant data to improve the properties of estimators of the standardized mean difference (SMD) effect size. Because samples recruited into education research studies are often more homogeneous than the populations of policy interest, the variation in educational outcomes can be smaller in these samples than…

This publication describes the methods and procedures used for the 2008/18 Baccalaureate and Beyond Longitudinal Study (B&B:08/18). The B&B graduates, who completed the requirements for a bachelor's degree during the 2007-08 academic year, were first surveyed as part of the 2008 National Postsecondary Student Aid Study (NPSAS:08), and then…

Likert-type item is the most popular response format for collecting data in social, educational, and psychological studies through scales or questionnaires. However, there is no consensus on whether parametric or non-parametric tests should be preferred when analyzing Likert-type data. This study examined the statistical power of parametric and…

This technical report provides detailed information about the sample, survey instruments, and resultant data for the Fall 2024 Pre-Kindergarten Teacher Survey (PKTS) which was administered via RAND's American Teacher Panel (ATP). The ATP is a nationally representative sample of public school teachers and part of the broader American Educator…

When performing an aggregate data meta-analysis of a continuous outcome, researchers often come across primary studies that report the sample median of the outcome. However, standard meta-analytic methods typically cannot be directly applied in this setting. In recent years, there has been substantial development in statistical methods to…

Extracting data from studies is the norm in meta-analyses, enabling researchers to generate effect sizes when raw data are otherwise not available. While there has been a general push for increased reproducibility in meta-analysis, the transparency and reproducibility of the data extraction phase is still lagging behind. Unfortunately, there is…

Moving Beyond the Ordinal Methodological Controversy: A Systematic Review (Manuscript 1): Ordinal outcome data is a common byproduct of education research. Yet more than seventy-five years after the development of Stevens' original measurement framework, the permissibility of select analytic techniques to ordinal outcome data remains a topic of…

'QuantCrit' (Quantitative Critical Race Theory) is a rapidly developing approach that seeks to challenge and improve the use of statistical data in social research by applying the insights of Critical Race Theory. As originally formulated, QuantCrit rests on five principles; 1) the centrality of racism; 2) numbers are not neutral; 3) categories…

In the contemporary society a massive amount of data is generated continuously by various means, and they are called Big-Data sets. Big Data has potential and limits which need to be understood by statisticians and statistics consumers, therefore it is a challenge to develop Big-Data Literacy to support the needs of constructive, concerned, and…

This paper emphasizes the divergences in quantitative and qualitative methodological approaches when exploring trust relationships in school units. We focus on how participants responded and how the results were interpreted. This study, based on such design, questionnaires, and semi-structured interviews to understand trust among teachers and…

What does it mean for a model to be a better model? One conceptualization, indeed a common one in Educational Data Mining, is that a better model is the one that fits the data better, that is, higher prediction accuracy. However, oftentimes, models that maximize prediction accuracy do not provide meaningful parameter estimates, making them less…

Providing a rich context has become a sine qua non of principled teaching of applied statistical thinking. With increasing opportunities to access secondary data, there should be increasing opportunity to work with rich context. We review the contextual information provided in 41 data sets suitable for introductory tertiary statistics teaching,…