Child Care and Early Education Research Connections

Skip to main content

The research glossary defines terms used in conducting social science and policy research, for example those describing methods, measurements, statistical procedures, and other aspects of research; the child care glossary defines terms used to describe aspects of child care and early education practice and policy.

Latent Growth Model
Latent growth modeling (LGM) is a class of statistical methods that are used to study change (growth) in behavior or attitudes over time. Traditional approaches to the study of change such as regression analysis and ANOVA focus on mean change (average change of a group or subgroups of study participants) and treat differences in change between participants as error. LGM examines individual (within-person) change over time as well as differences in the individual change (between-person). It is used to model change over time and to investigate factors that affect the level and rates of change. It can be used to examine differences in which groups of people change.
Latent Variables
In statistics, latent variables are variables not directly observed and measured but inferred from other observed and measured variables. Mathematical models (e.g., factor analysis, structural equation modeling, item response theory models) are used to examine the relationships between a set of observed variables (indicators) in order to identify the latent variable. For example, the latent variable 'teacher attitudes toward math' may be modeled from a series of survey items asking about their feelings toward math and how they feel when doing math.
Least Squares
A commonly used method for calculating a regression equation. This method minimizes the difference between the observed data points and the data points that are estimated by the regression equation.
Level of Significance
See significance level.
Likert Scale
A Likert Scale is a type of rating scale used to measure attitudes, values, or opinions about a subject. Survey respondents are asked to indicate their level of agreement or disagreement with a series of statements. The responses are often scaled and summed to give a composite measure of attitudes or opinions about a topic.
Limited Dependent Variable
A limited dependent variable is a variable with the range of possible values "restricted in some important way." Examples include binary variables that have only two values (e.g., child attends child care or not; child is promoted to next grade or not). Also, variables that can only take on certain values (e.g., discrete variables that have a limited set of categories or continuous variables that can only have positive values such as hours worked or wages earned).
Linear Regression
A statistical technique used to find a linear relationship between one or more (multiple) continuous or categorical predictor (or independent) variables and a continuous outcome (or dependent) variable.
Literature Review
A comprehensive survey of the research literature on a topic. Generally the literature review is presented at the beginning of a research paper and explains how the researcher arrived at his or her research questions.
Local Average Treatment Effect
The local average treatment effect (LATE) is the average effect of a treatment for the group of individuals who complied with the random assignment. That is, it is the estimated effect when those assigned to the treatment group received the treatment and those assigned to the control group did not.
Log-Linear Analysis
A technique used in statistics to examine the relationship between two or more categorical variables. The technique is used to test whether the variables are independent of one another or associated in some way. For example, a log-linear model might be used to test whether there is an association between children being 'red shirted' (children whose parents hold them back from entering kindergarten even though they meet the age eligibility requirement for kindergarten entry) and their sex (boys versus girls) and race/ethnicity (African American, Hispanic, White, Asian). The model would test for both main effects and interaction effects.