Seminars on Quantitative and Survey Methods

Facilitated by Dr. Rong Chen For doctoral students who are interested in conducting rigorous quantitative dissertation studies, it is critical to: be familiar with the secondary datasets that are available and how to access them, know how to craft potential research questions from relevant data, and identify the appropriate methods and strategies for the selected research questions. Such knowledge and skills will also help prepare students to become competent researchers or administrators in educational, governmental, and other agencies concerned with educational measurement, statistics, and program evaluation. This seminar is intended to introduce students to the range of large-scale national and international data sources for quantitative studies and to provide examples of how these datasets have been used in educational research. By the end of the seminar, students will be able to understand where some important large-scale survey databases in education are available, the major research issues that may be addressed with these resources, and the advantages as well as challenges/strategies of using large-scale data. In addition, we will look at some example empirical studies that have utilized such data in educational research.

Facilitated by Dr. Paul Garton Questions in educational research tend to have causal implications. What is the effect of some reading program on reading levels? Do private school vouchers cause public schools to change in quality? Direct cause-and-effect relationships are often a topic of interest, because we want to know how to enact a change or identify the (un)intended results of an existing program. Causal claims, however, require specific research designs and analyses. The gold standard of causal research is the experimental design, in which the researcher has full control over which participants receive treatment, but different methods can be used to approximate experimental designs without having control over treatment assignment. This seminar introduces the basic intuition behind what the scholarly community currently understands as causal designs, clarifies what it means for a study to be unable to make causal claims, and gives an overview of the most common approaches to causal analysis, difference-in-differences and regression discontinuities.

Facilitated by Dr. Manuel Gonzalez In this seminar, Dr. Gonzalez will provide an applied discussion about factor analysis in survey research. The discussion will cover topics such as (a) why researchers use multi-item survey measures, (b) measuring unidimensional versus multidimensional concepts, (c) the difference between exploratory and confirmatory factor analyses (EFA and CFA), (d) factor rotation, and (e) how to select the right factor analysis strategy for your research. The seminar will conclude with a STATA tutorial using real-world survey data, followed by an open question and answer session with the audience.

Facilitated by Dr. Jason Burns Surveys can be a valuable means of gathering data 1) when the phenomena of interest are not directly observable, such as with satisfaction, professional commitment, and issues around culture and climate, and/or 2) when existing datasets do not contain the information needed to answer one’s research question(s). However, there are many ways that error, or inaccuracy, can be introduced during the process of conducting a survey. For instance, respondents may not understand an item, respondents may interpret the language used in an item in different ways, or the wording of items may steer respondents toward a particular answer. One way to reduce error in survey data is to employ a rigorous process when developing survey items - the set of questions that will be asked to respondents. This seminar provides an overview of the process of constructing high-quality survey items. Topics covered will include survey terminology, the characteristics that define high-quality survey items, common types of “poor” survey items, and how to evaluate items before using them in analysis.

Facilitated by Dr. Robert Kelchen Most quantitative research studies employ a form of regression analysis to examine the relationship between some dependent, or outcome, variable and some independent, or predictor, variable(s). While this often seems straightforward at first, employing regression is sometimes complicated by the fact that datasets often contain dozens, or sometimes even hundreds, of variables. Answering research questions using regressions requires the researcher to consider the characteristics of the variables at hand (Are they string, binary, categorical, or continuous?) as well as how they may be related (Does theory posit a relationship? What are the conventions used in other studies of the topic?). This seminar addresses these issues, focusing on 1) identifying the type(s) of regression that are appropriate for a study 2) selecting the relevant dependent and independent variable(s) for a regression model, and 3) running a regression model in Stata.

Facilitated by Dr. Jason Burns Quantitative researchers often find that a significant component of their work involves preparing data for analysis, such as by recoding or transforming variables, creating new variables, adding variables/columns to the data from other sources, and adding observations/rows to the data. This is especially the case when working with secondary data – data collected by others such as a state department of education or the federal government. The process of preparing data for analysis can be time-consuming, but many aspects of data cleaning and management can be automated using software such as Stata. This seminar focuses on the following aspects of data cleaning and management in Stata: 1) recoding variables 2) creating and transforming variables 3) adding new variables/columns from a different dataset, and 4) adding observations/rows from a different dataset.

Facilitated by Drs. David Reid and Jason Burns The COVID-19 Pandemic impacted primary, secondary, and postsecondary education in multiple ways. Correspondingly, researchers and practitioners have understandably taken great interest in research that explores the way(s) in which the pandemic shaped how educators provided instruction, student outcomes, and how leaders worked to build and maintain community, among other things. While understanding the effect(s) of the pandemic are of great importance to practitioners and policymakers, there are some important methodological challenges that can impose limits on the extent to which we can fully understand the true impact of the many facets of the pandemic. In this seminar, Drs. Reid and Burns will highlight the factors that researchers will need to consider when designing studies that focus on the pandemic so that valid conclusions can be drawn. This seminar is especially relevant to doctoral students who are considering a dissertation that in some way touches on the effects of the pandemic on students, educators, and leaders.

Facilitated by Dr. Christopher Tienken A common challenge for graduate students and researchers is in setting a research agenda: identifying a topical area of study and using that to generate a set of research questions that guide future inquiry. In doing so, one must strike a balance between formulating a topic that is narrow enough that it can be studied, but still broad enough that it is relevant to a broader audience. This seminar focuses on practical tips one can use to develop a research topic and discusses how this interacts with the frameworks and methods one may use to explore it.