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.
A approach to conducting research that involves collecting, analyzing and integrating data from quantitative (e.g., experiments, surveys) and qualitative (e.g., focus groups, observations, qualitative interviews) methods. The goal is to provide greater breath and depth of understanding of a topic.
A descriptive statistic that is a measure of central tendency. It is the value that occurs most frequently in the data. For example, if survey respondents are ages 21, 33, 33, 45, and 76, the modal age is 33.
A moderating variable, also called a moderator variable, is a variable that affects the strength or direction of the relationship between a dependent and independent variable. It is represented by an interaction term in regression models. For example, a researcher might examine whether maternal depression moderates the relationship between the quality of mother-child interactions and children's development. The researcher might hypothesize that this relationship will be weaker for children's whose mothers are depressed than for other children.
A form of average which has been adjusted (or "smoothed") to allow for seasonal or cyclical components of a time series.
Multidimensional scaling (MDS) is a tool used by researchers to identify and quantify the relationships among the responses of study participants to a group of items in a survey or other data collection instrument. The outcome of MDS is a visual representation of the patterns of similarities, dissimilarities and distances among these participants.
Experiments that are conducted at multiple locations, often by several research organizations. Multi-site studies provide larger samples, which increases the statistical power to detect significant treatment effects. Given differences in the populations in the different communities, samples are often more diverse, which can increase the likelihood that the effects are not solely limited to a single population.
Multistage sampling is a probability sampling technique where sampling is carried out in several stages. It is often used to select samples when a single frame is not available to select members for a study sample. For example, there is no single list of all children enrolled in public school kindergartens across the U.S. Therefore, researchers who need a sample of kindergarten children will first select a sample of schools with kindergarten programs from a school frame (e.g., National Center for Education Statistics' Common Core of Data) (Stage 1). Lists of all kindergarten classrooms in selected schools are developed and a sample of classrooms selected in each of the sampled schools (Stage 2). Finally, lists of children in the sampled classrooms are compiled and a sample of children is selected from each of the classroom lists (Stage 3).
A situation in which two or more predictor (independent) variables in a sample are highly related to each other. When using regression analysis, this can lead to incorrect estimates of their individual effects on the outcome (dependent) variable. Multicollinearity violates an underlying assumption of regression that each predictor (independent) variable has an independent impact on the outcome (dependent) variable.
Multilevel data are organized at more than one level. That is, the data are nested. In research that involves children enrolled in early childhood programs such as Head Start, the children in the study are often nested in classrooms, which are nested in Head Start centers.
A model involving variables measured at more than one level of a hierarchy. An obvious hierarchy consists of children nested in classes, and classes nested in schools. Measurements can be obtained for child characteristics, class and teacher characteristics, or school characteristics. Multilevel models are also known as hierarchical linear models or random coefficient models. Multilevel are use to solve the statistical problems caused by dealing with hierarchically nested data.