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.
Influences that distort the results of a research study.
A distribution in which two scores or values are the most frequently occurring. Interpreting the average of a bimodal distribution is problematic because the data are not normally distributed. Identifying bimodal distributions is done by examining a frequency distribution or by looking at indices of skew or kurtosis, which are frequently available with statistical software packages.
A popular method for variance estimation in surveys. It consists of subsampling from the initial sample. Within each stratum in the sample, a simple random subsample is selected with replacement. This creates a finite number of new samples (or repetitions). The same parameter estimate is then calculated for each of the subsamples. The variance of the estimated parameter is then equal to the variance of the estimates from these subsamples.
Canonical Correlation Analysis
Canonical correlation analysis is used to examine the associations between multiple independent variables and two or more intercorrelated dependent (outcome) variables. That is, it is used in situations where multiple regression would be used, but where there are multiple dependent variables that are correlated with each other. For example, a researcher might be interested in the associations between children's race/ethnicity, gender and family SES and their performance of several measures of academic achievement (e.g., math, reading, science).
An intensive investigation of the current and past behaviors and experiences of a single person, family, group, or organization.
Variables with discrete, non-numeric or qualitative categories (e.g. gender or marital status). The categories can be given numerical codes, but they cannot be ranked, added, multiplied or measured against each other. Also referred to as nominal data.
Categorical Data Analysis
Categorical data classify responses or observations into discrete categories (e.g., respondents' highest level of education is often classified as less than high school, high school, college, and post-graduate). While there are many techniques for analyzing such data, 'categorical data analysis' usually refers to the analysis of one or more categorical dependent variables and the relationships to on or more predictor variables (e.g., logistic regression).
An analysis that seeks to establish the cause and effect relationships between variables.
Causal Inference with Interference
One assumption of randomized experiments is that a subject's response to treatment (intervention) depends only on the treatment to which the subject is assigned, not on the treatment assignments of other subjects. The ability to draw causal inferences from the findings of experimental studies rests on the assumption that there is not such interference. However, such interference is common in experimental studies where there is a high level of interaction between subjects, such as studies of different curricula where teachers in the same school are assigned to different treatment groups.
The highest limit of performance that can be assessed or measured by an instrument or process. Individuals who perform near to or above this upper limit are said to have reached the ceiling, and the assessment may not be providing a valid estimate of their performance levels.