psy 227 - final exam

Linear regression

This procedure is used as an extension of correlation. It is used only when two variables are found to be a least fairly strongly correlated. A researcher may be interested in using the strong relationship between the two variables to be able to predict scores on one of the variables in the future based on scores that she collects on the other variable. If so, then she would calculate the regression equation. The equation will tell her the predicted value of the Y variable based on the known value of the X variable.

 

Two independent samples design/ t-test for independent samples

This type of design involves collecting data from two samples, each one representing one condition of the one independent variable. The two samples must be independent from one another. This means that subjects were either randomly assigned to one condition or the other or subjects were assigned to one of the two conditions based on some preexisting state that made them fit one condition or the other of  the independent variable. There is one dependent variable, which must be measured on either an interval or a ratio scale.

Two matched samples design/ t-test for related samples

 

This type of design involves collecting data from two samples, each one representing one condition of the one independent variable. The two samples must be related to one another in that each participant in condition one has been carefully matched to one participant in condition two. The participants are matched on important characteristics that are thought to have some relationship to the dependent variable such as race, age, gender, or socioeconomic class. There is one dependent variable, which must be measured using either an interval or a ratio scale.

Two sample repeated measures/t-test for related samples

This type of design involves collecting data from two samples, each one representing one condition of the one independent variable. The two samples must be related to one another in that each participant in condition also serves as a participant in condition two. This type of design usually involves presenting one condition of the independent variable to the group of participants and then measuring the dependent variable. Then the group is allowed to return to baseline. Once baseline has been returned to, the second condition of the independent variable is presented to the same group and, again the dependent variable is measured. This type of design is also used for pretest/posttest evaluations where a group is measured before a treatment is given, and then again, after the treatment is given. The pre-test is then compared to the post-test score. The one dependent variable must be measured using either an interval or a ratio scale.

Between-groups design for more than two samples/ One-Way Analysis of Variance (ANOVA) for independent samples:

This type of design involves collecting data from more two independent samples, each one representing one condition of the one independent variable. All samples must be independent from one another. This usually means that each subject was randomly assigned to one of the conditions of the independent variable. This procedure allows us to compare all conditions of an independent variable at the same time. The one dependent variable is measured using either an interval or a ratio scale.

The Between groups factorial design for two independent factors/ Two –Way Analysis of Variance (ANOVA) for independent samples

This type of design involves collecting data from at least two samples for each of two independent variables. Each sample represents a combination of one level of independent variable A and one level of independent variable B. All samples must be independent from one another. This usually means that each subject is randomly assigned to one of the AXB cells. This design allows us to compare all levels of two independent variables at the same time, and also evaluate whether there is an interaction between the levels of the two independent variables. The one dependent variable is measured using either an interval or a ratio scale.

The chi square for equal expected frequencies

This is a nonparametric statistical procedure that allows us to compare the frequencies that we observe in different categories of the independent variable to test whether they are equal or not. This type of procedure is used with nominal data where there is a minimum frequency of 5 for each category.

The chi square for unequal expected frequencies

This is a nonparametric statistical procedure that allows us to compare the frequencies that we observe in different categories of the independent variable to test whether the distribution of observed frequencies matches the distribution of some known or hypothesized population. This type of procedure is used with nominal data where there is a minimum frequency of 5 for each category.

Dr. Lopez is studying whether people prefer certain types of pets in specific proportions. She predicts that 50% of people prefer dogs, 30% prefer cats, and 20% prefer other animals. She surveys 300 participants and records their preferences. The dependent variable is the number of people who prefer each type of pet.

One-way Chi Square (unequal expected frequencies)

Dr. Kim wants to know whether college students choose among four majors (psychology, biology, business, and engineering) equally. She surveys 200 students and records how many choose each major. The dependent variable is the number of students in each category.

One-way Chi Square (equal expected frequencies)

Dr. Patel is studying the effect of caffeine on memory. He randomly assigns 100 participants to either a caffeine group or a no-caffeine group. After consumption, all participants complete the same memory test. The dependent variable is the memory test score.

t-test for independent samples

Dr. Nguyen is studying the effect of exercise on mood. She measures participants’ mood scores before starting a 4-week exercise program and again after the program ends. The dependent variable is mood score.

t-test for related samples (repeated measures)

Dr. Rivera is studying the effect of sleep on reaction time. She pairs participants based on age and baseline reaction time. One participant in each pair is assigned to a sleep-deprived condition, and the other gets a full night of sleep. The dependent variable is reaction time.

t-test for related samples (matched samples)

Dr. Chen is studying how study method and environment affect exam performance. Participants are randomly assigned to one of four groups: (1) studying alone in silence, (2) studying alone with music, (3) studying in a group in silence, or (4) studying in a group with music. The dependent variable is exam score.

Two-way between-subjects ANOVA

Dr. Brooks is studying whether the number of hours students study predicts their exam scores. She collects data on how many hours each student studies per week and their final exam score. The dependent variable is exam score.

Linear Regression

Dr. Singh is studying the effect of teaching style on student performance. Participants are randomly assigned to one of three groups: lecture-based learning, interactive learning, or online learning. The dependent variable is test score.

One-way between-subjects ANOVA