Answers to Test Yourself
(1) This is a correlational study. No independent variable was manipulated, and other possible factors that could affect cancer rates were not controlled for. Thus, we cannot conclude that drinking green tea has a causal relationship with cancer rates. It is possible that it does, or it is possible that people who are less likely to get cancer (e.g., no family history) are more likely to drink green tea (i.e., low cancer probability causes one to drink green tea). Finally, it is possible that a third variable (e.g., maintaining a healthy lifestyle) causes one to both drink green tea and have low likelihood of getting cancer. (2) An experiment would need to be conducted where the researcher randomly assigns participants either to drink green tea every day or to not drink any green tea (but perhaps to drink a similar-looking drink that they are told is green tea but is actually just colored water) and then records who gets cancer in order to compare cancer rates across the two groups. (3) A descriptive research question is one where a researcher is interested in whether a behavior occurs and how often it occurs. In a correlational study, a descriptive research question might be whether a relationship exists between two variables and what type of relationship exists. A predictive research question is one in which a researcher examines whether one variable can be predicted from one or more variables. (4a) Weather as measured by temperature that day and barometric pressure that day, cognitive abilities, mood as measured by a questionnaire, activity level, and time spent outside that day. (4b) narcissism as measured by a survey, explicit self-esteem as measured by a survey, and implicit self-esteem as measured by the IAT. (4c) predictor variables--family stress and functioning measured by surveys, outcome variable--expressed emotion measured by a survey. (4d) predictor variable--expectations about counseling measured by a survey before therapy, outcome variable--client–therapist working alliance measured by a survey after therapy. (5a) How often people exercise measured by how many times per week they exercise for 30 minutes or more at a time and happiness measured by the score on the quality of life scale. (5b) more exercise may cause people to be happy, being happy could cause people to exercise more, or some third variable could cause both happiness and more exercise. (5c) no variable is manipulated in this study; thus, no causal relationships are tested. One of the variables (it would be easiest to manipulate the amount of exercise variable) must be manipulated (e.g., participants randomly assigned to different conditions of the variable) to test a causal relationship. (6) positive. (7) predictive. (8) Correlational studies measure dependent variables from a single group of subjects to look for relationships between the measures, but do not separate subjects into groups for comparison as is done in quasi-experiments.