Titlebook: Statistical Methods for Experimental Research in Education and Psychology; Jimmie Leppink Textbook 2019 The Editor(s) (if applicable) and

JEER

ventilate

Dealing with Missing Dataith other types of research and large proportions of missing data are uncommon in most of experimental educational and psychological research, the topic is equally relevant to experimental research. Frequently encountered methods for dealing with missing data are mean imputation, listwise deletion,

付出

笨重

Multicategory Nominal Outcome Variablesh this type of outcome variable remains less common than dichotomous, ordinal (Chap. .) or quantitative outcome variables (Chap. .), it may constitute a main outcome variable in for example experiments that focus on choice behaviour or association as a function of treatment. For instance, in researc

neurologist

GLADE

Quantitative Outcome Variablesan example with a quantitative outcome variable and no substantial departures from assumptions is already covered in Chap.?., this chapter focusses on three types of somewhat more difficult situations: considerable skewness in a time outcome variable, skewness inherent to the nature of a count outco

間諜活動

Common Approaches to Multiple Testingts, with the exception of Experiment 3 in Chap. . with ANOVA and polynomial regression analysis on a five-groups treatment factor. When a treatment factor consists of more than two groups but specific hypotheses about group differences are lacking, applying a correction for multiple testing to keep

棲息地

Directed Hypotheses and Planned Comparisonseses being tested being true (no matter how large that series) and not having specific hypotheses that would justify focusing on only a few instead of on all comparisons. When a treatment factor consists of more than two conditions but—based on theory, previous research or common sense—we do have sp

FOIL

Flu表流動

Factor-Covariate Combinations on a categorical or quantitative outcome variable. However, in not so few cases, we have at least one other, non-treatment variable measured along with our primary outcome variable of interest. How to deal with this additional variable depends on whether it is measured before or after the start of