Statistics, Tests, And Measurements Flashcards
Descriptive studies
Type of research design that gathers data and presents a complete picture of a given subject.
Case study, survey
Confounds
Factors that could explain the results but that are not directly measured or addressed by the study.
Correlational studies
Type of research design that tries to figure out the relationship between two or more variables and that grades this relationship with the Pearson Correlation Coefficient (-1; 0; 1).
Correlation is not causation!
Experimental research
Type of research design that determines the causation between variables.
Reliability
Your research is reliable when your tools for measuring a variable do it accurately and consistently.
Validity
The legitimacy of the research and its conclusions.
Construct validity: when measurements correspond to question
Internal validity: the confirmation of the reality of the causal relationship of your variables
External validity: conclusions applied to more people
Statistical analysis
Used to check whether or not the data supports or rejects the hypothesis.
Descriptive statistics: describe and summarize a set of data
Inferential statistics: used to draw conclusions from the data described by descriptive statistics. Statistical significance is confirmed when there is a less than 5% chance for the results to occur due to chance.
Factor analysis
Factor analysis is a statistical method that is used to determine whether a group of observable variables are related to a smaller group of underlying factors. CFA and EFA are the two types of factor analysis. The differences between CFA and EFA are as follows.
CFA requires you to predetermine: a specific hypothesis based on previous research or theory; the number of factors; which observable variables are related to each factor.
EFA: does not require you to predetermine the number of factors or the relationship between the factors and the observed variables; identifies the factor structure; can explain a maximum variance amount.
