Statistics How To

Experimental Design > Replication and Replicability Contents: Replication Replicability What is Replication? In experimental design, replication is where each treatment is assigned to many participants. Or, the entire experiment is repeated on a large group of subjects. The process: Improves the significance of your experimental results. Reduces variability . Increases confidence in your results. If … Read more

Types of variables > An antecedent variable (or antecedent confounding variable) happens before the independent variable and dependent variable. It may affect the relationship between the two variables, or it might just affect the two variables separately, without their being a direct relationship between the IV and DV (this is called a spurious relationship). Examples … Read more

Statistics Definitions > In psychology, ipsative measurement refers to a particular type of survey or assessment in which a respondent is forced to choose among several generally desirable options. Unlike normative tests, where a respondent answers questions with scaled responses, ipsative tests do not have an scale within the question. While normative tests can compare … Read more

Statistics Definitions > What is Cross Validation? Cross validation (also called rotation estimation, or out-of-sample testing) is one way to ensure your model is robust. A portion of your data (called a holdout sample) is held back; The bulk of the data is trained and the holdout sample is used to test the model. This … Read more

Probability > Objective Probability What is Objective Probability? Objective probability (also called frequency probability) is associated with random events like die rolls, choosing bingo balls, or your numbers coming up in the lottery. The term is mostly used in philosophy as a philosophical interpretation of probability, rather than a process itself. The term “objective probability” … Read more

Statistics Definitions > What is a Representative Sample? A representative sample is where your sample matches some characteristic of your population, usually the characteristic you’re targeting with your research. For example, if you’re conducting a survey about how women’s experience as a data scientist compare to men, you would want your sample to reflect the … Read more

Reliability and Validity > Composite reliability (sometimes called construct reliability) is a measure of internal consistency in scale items, much like Cronbach’s alpha (Netemeyer, 2003). It can be thought of as being equal to the total amount of true score variance relative to the total scale score variance (Brunner & Süß, 2005). Alternatively, it’s an … Read more

Types of Variables > Research Variable / Study Variable A research variable (also called a study variable) is an informal term that means any variable used in research that has some kind of cause and effect relationship. A research/study variable can be one of a wide variety of variables used in a study, including independent … Read more

RCTs > Intervention Fidelity in Research answers the question “Was the intervention delivered as intended?” It’s a measure of how well the results of an intervention or experiment matched up to what was actually planned. An intervention is some form of treatment, but it could also be a specified active or non-active non-treatment. For example, … Read more

Nearest Neighbor matching > k-NN (k-Nearest Neighbor) K-nn (k-Nearest Neighbor) is a non-parametric classification and regression technique. The basic idea is that you input a known data set, add an unknown, and the algorithm will tell you to which class that unknown data point belongs. The unknown is classified by a simple neighborly vote, where … Read more