Part 1: Making Hypotheses

For each study described below, state the null hypothesis. Then review the results for the study and decide what your decision (accept or reject) should be with regard to the two hypotheses. Remember, you should NEVER accept a null hypothesis.

Study 1

Alternative hypothesis: Anxiety increases lying in children.

a. Null hypothesis:

Study 1 results: A study compared 50 children who were placed in an anxiety-inducing situation with 50 children in a control group. They were then asked about their behavior, and the number of inaccurate responses was recorded. Inaccurate responses did not differ for the two groups.

b. We should ______________ the null hypothesis.

c. We should ______________ the scientific hypothesis.

d. Suppose that the study described above was conducted inaccurately and that anxious children really do lie more. In this case, we have made a

Type I

or

Type II decision error.

Study 2

Alternative hypothesis: People are more likely to help a stranger if there is no one else around than if they are in a group (i.e., the bystander effect).

a. Null hypothesis.

Study 2 results: A situation was set up on a busy highway where someone needed help with her car. Researchers observed 100 cars drive by and counted the number of people who stopped and whether they were alone or with other people in the car. People driving with others stopped less often to help than people who were driving alone.

b. We should _________________ the null hypothesis.

c. We should _________________ the scientific hypothesis.

d. Suppose that the above study contains a confounding variable, and when tested properly, people driving with others actually stop just as often as people driving alone. In this case, the study described above has led us to make a

Type I

or

Type II decision error.

Whenever we reject the null hypothesis, what does this tell us about the independent variable?

Part 2: Inferential Statistics

Study 3

A behavioral psychologist conducts an experiment to determine whether operant conditioning techniques can be used to improve balance in people who consider themselves "clumsy." She recruits 50 participants, each of whom responds yes to the question "Do you consider yourself clumsy?" on a pre-experimental questionnaire. Half the participants are given a balance task (stand on one foot with your arms in the air) with the time they can balance recorded. They are then excused and asked to return 3 weeks later. The other half of the participants are given 3 weeks of operant training during a balance exercise. In this training, the participants are asked to perform the balance task described above. Each time they can beat their previous balance time, they receive $10. After the 3-week period. all participants are asked to perform the balance task again. Alpha is set at .05. The members of the training group can balance for an average of 25 seconds. Members of the control group (who did not receive the training) can balance for an average of 24.3 seconds. When the inferential statistical test is conducted on these

data, p = .08.

1. What is the independent variable for this study? The dependent variable?
2. What is the alternative hypothesis?
3. What is the null hypothesis?
4. What population is being tested in this study?
5. Is the difference observed in the study statistically significant? Why or why not?
6. Based on your answer to (5) above, what decision should be made about the null hypothesis?
7. In the context of this experiment, what exactly is p the probability of?

Study 4

Some have claimed that children of divorced parents are not as well adjusted as children of parents who are married. To test this idea, you set up a study with two groups. One group consists of 100 children who have divorced parents. The other group consists of 100 children who have married parents. Each child in the study is asked to complete the Social Adjustment Scale for Children (SAC). Higher scores on the scale indicate better social adjustment.

1. Does this study contain an independent variable? Why or why not?
2. What dependent variable is measured in this study?
3. What is the null hypothesis for this study?
4. What is the alternative hypothesis?

The results indicate a mean difference of 10 on the SASC scale between the two groups of children. With alpha

set at .05, p = .02.

1. What decision should be made with regard to the null hypothesis?
2. Describe one possible source of bias that could be present in this study causing the results that were observed.