Gustafson, R. (1986c) Threat as a determinant of alcohol-related aggression. 58:287-297. [Investigates the hypothesis that threat will inhibit aggression in intoxicated subjects. Results confirmed the hypothesis.]

Hypothesis and options are used in the description of beliefs.

This is just a brief example of scientific method.

Causal comparative research ckv - SlideShare


No manipulation Individual variable is manipulated

Weaker evidence for causation Stronger evidence for causation

Groups already formed Can assign subjects to groups

Researchers observe what HAS happened Researcher can design experiment Associations Between Categorical Variables Categorical variables make it possible to investigate the association between variables using cross break tables and contingency coefficients.

Causal comparative research - SlideShare

The quantitative research design that we select subsequently determines whether we look for relationships, associations, trends or interactions. To learn how to structure (i.e., write out) each of these three types of quantitative research question (i.e., descriptive, comparative, relationship-based research questions), see the article: .

An ex post facto, causal-comparative design was employed


2 or more groups Score for each variable for each subject

1 categorical variable (group member) 2 or more quantitative variables

Analyze data by comparing averages Analyze data using scatterplots and/or
or using crossbreak tables correlation Similarities

At least 1 categorical variable

Compare group performances (average scores) to determine relationships

Typically compare separate groups of subjects



 What is causal-comparative research

Thus B and D (local organization and weak state institutions) turn out to be necessary conditions for the occurrence of revolution (in this hypothetical world)-a circumstance that can be verified by inspection of the causal diagram directly.

-causal-comparative research is interested in ..

Whilst we refer to this type of quantitative research question as a relationship-based research question, the word relationship should be treated simply as a useful way of describing the fact that these types of quantitative research question are interested in the causal relationships, associations, trends and/or interactions amongst two or more variables on one or more groups. We have to be careful when using the word relationship because in statistics, it refers to a particular type of research design, namely experimental research designs where it is possible to measure the cause and effect between two or more variables; that is, it is possible to say that variable A (e.g., study time) was responsible for an increase in variable B (e.g., exam scores). However, at the undergraduate and even master's level, dissertations rarely involve experimental research designs, but rather quasi-experimental and relationship-based research designs [see the section on ]. This means that you cannot often find causal relationships between variables, but only associations or trends.

Empirically testing a hypothesis is the process of ..

A sample size of 411; satisfactorily large.
Questionnaires looking at preferred social context, social ability and preferred therapy
One Half sample sizes used paper and the other half computer.
Internet abuse based on preexisting criteria for substance abuse
Changed a quantitative to a categorical variable
Independent Variable – Internet use
Dependent variable - Depression and social dysfunction
Procedures/results and Internal Validity
Used three research methods; survey, casual comparative, and correlation with respect to relationships among variables
Internal Validity – males were more likely to be Internet abusers than females
Conclusion: Too much reliance of inference techniques, the validity of the Instrument with using a self-reporting scale

role of causality in the English language, confirming our hypothesis.

According to their hypothesis, as arousal levels increase, so do the performance levels of the individual, but only to certain point (known as the ‘optimal’ point)....