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.
Narrative research is a group of qualitative approaches that rely on the written or spoken words or visual representation of individuals to understand how people create meaning in their lives. The focus is most often on the lives of individuals as told through their own stories. The emphasis is on the story, what it says and how it is told.
A natural experiment is an empirical study in which individuals or groups of individuals are sorted by "nature" (i.e., by conditions or events that were not controlled by the researcher) into something like treatment and control groups. They differ from Randomized Controlled Trials (RCTs). The key assumption of a natural experiment is that there are comparable groups, and one group is randomly affected by forces outside their control ("nature"). For example, researchers often study the effects of nature disasters on a range of outcomes.
Negative Binomial Regression
Negative binomial regression is a type of generalized linear model that researchers use when the dependent variable is a count of the number of times an event occurs. For example, a researcher might use negative binominal regression to study the relationships between several variables (e.g., children's general health status and type of early care and education program) and the number of school days children have missed during the school year.
See categorical data.
A scale that allows for the classification of elements into mutually exclusive categories based on defined features but without numeric value.
A nonlinear model describes nonlinear relationships between the dependent and independent variable(s). A linear model assumes that the dependent variable changes by a fixed amount for each unit change in the independent variable. A nonlinear model, on the other hand, does not make this assumption. Instead of the relationship between the dependent and independent variable being represented by a straight line, a nonlinear relationship is characterized by one or more curves.
When analyzing time series data, a linear trend is one where the data increase by a constant amount at each successive time period. A linear trend is represented by a straight line. However, data do not always increase by the same amount. For example, data that increase by varying amounts at each successive time period show a nonlinear, curvilinear trend.
Nonparametric statistics refer to the group of statistical methods that require fewer assumptions about the distribution of the data. For example, nonparametric tests of significance such as the Chi-square test dos not require the data to fit a normal distribution. Nonparametric statistical methods are used when analyzing nominal, ordinal or ranked data.
A type of error that is caused when a portion of the sample with particular characteristics do not respond to a survey. For example, individuals who are trying to dodge bill collectors might be less likely to answer their telephone and therefore may be less likely to respond to a telephone survey. This could lead to biased statistical results because individuals who do not pay their bills would be less likely to answer the survey. Researchers try to correct for this problem by determining the characteristics of those who were less likely to answer the survey (for example, they might be younger and have lower incomes) and controlling for those characteristics in the analysis, by imputing missing data, or by using sampling weights (analysis weights) that adjust for nonresponse.
Nonresponse Rate Bias
A source of bias that occurs when non-respondents differ in important ways from respondents.