Each question describes a research study. You must decide what is the appropriat

Question

Each question describes a research study. You must decide what is the appropriate way to

analyse the results. For each, state the statistical procedure you would use and why it is the

most appropriate. A good answer would take 1-3 sentences. No calculations required.

1. You are investigating the influence of social activities on life satisfaction. You have 3

conditions, each with 100 randomly-assigned participants. In each condition participants

are assigned to engage in a different social activity 1 evening a week for a month. The

conditions are 1) spend the evening browsing friends Facebook time lines, 2) spend the

evening having dinner with close friends, 3) spend the evening having dinner with

immediate family members. After a month, everyone competes a standardized life

satisfaction inventory (higher scores = more satisfied). What hypothesis test procedure

would you use to determine if there are any significant differences between the treatment

conditions?

2. You want to know if chess skill is related to the age when a player first learned the

game. You send a survey to the top 100 ranked chess players asking them how old they

were when they first played chess. For each player you end up with two number: their

rank and their age of first-play. How would you decide if these numbers are related?

3. You are interested in the effects of two independent variables on the dependent variable

of depression symptoms. The independent variables are dose of a new antidepressant

drug and gender. Fifty males and 50 females are included in your study and 100mg and

200mg of the antidepressant drug are divided evenly between the participants and taken

for one month. Participants rank their symptom severity after the month. How would the

effect of the antidepressant drug and gender be measured?

4. You want to investigate the effect of violent video games on aggressive behavior. 57

participants are first asked to take part in a task designed to measure aggression based on

how long (in seconds) they hit a punching bag. The participants then play a violent video

game for 3 hours. Then the participants come back to the lab about 24 hours later to

participate in the aggression task a second time.

5. You are interested in whether a person’s level of extraversion predicts how many dates

they go on in college. Extraversion is measured using a personality assessment on a

continuous scale ranging from 1-7. The number of dates are reported by participants as

the total sum of dates they’ve been on from the start of their freshman year to the time

they graduate. You want to predict how many dates someone with a extraversion score

of 6 would go on.

6. You want to compare the mathematical abilities of Michigan State University

undergraduate students on a standardized test (total score out of 100) to the national

average. You take a sample of MSU students, administer the test, and receive a mean

score for the sample ( M = 79). You want to compare that to the established national

average ( = 74, = 3).

7. A professor is curious as to how her class of n=122 is currently performing compared to

past sections of the course she has taught at the university. She compares her classes

current mean grade ( M =81, s =4) to past averages at this point in the semester =77.

Which hypothesis test needs to be conducted in order for the professor to determine if her

current class is performing better than past classes using a two-tailed test with =.05?

8. You want to know if apologies have a significant effect on individuals who are

interrupted while playing a game. You take a sample of 100 students and have them play a

computer game that intentionally crashes as they progress. Half of the participants receive

an apology and the other half do not. Afterward, all participants rate their game playing

experience and state whether they received an apology from the game or not. How would

you decide if there is a significant effect?

9. A psychologist is interested in testing whether political party affiliation relates to whether

or not people vote in local elections. She decides to do a study in which she randomly

samples 300 individuals and for each person she determines political party affiliation

(Democrat, Republican, Independent) and she also asks whether or not the person voted in

their most recent local election. What is the most appropriate statistical test to answer the

researcher’s question?

10. A researcher wants to test whether reward size affects the number of trials it takes for a

child to learn a complex task. He decides to use 4 types of reward (very high value, high

value, moderate value, low value). The researcher gets a random sample of 40 4 year old

children and she randomly assigns each child to one of the four reward conditions. She then

runs through the training task with each child as many times as she needs to for the child to

be able to do it successfully on his/her own. The outcome variable is thus the number of

times it takes for the child to succeed. What is the most appropriate statistical test to answer

the researcher’s question?

Explanation / Answer

Question 1

For the first scenario, we have to compare the significant differences between the three treatment conditions. We have to find whether there is any significant difference in the average satisfaction scores between the three treatment conditions or not. For checking this significant difference, we need to use the one way analysis of variance. This means we will use the single factor ANOVA F test for finding the significant difference in the three average satisfaction scores.

Question 2

For the given scenario, we want to find out whether there is any significant relationship exists between the age of first play and the rank of chess player. This means we have to find out the correlation coefficient between the two variables age of first play and rank of chess player and then we have to check the hypothesis or claim whether the variables age of first play and rank of chess player are statistically related to each other or not. For checking this we have to use the t test for checking the significant correlation or linear relationship exists between the two variables.

Question 3

For the given scenario, we have to find out whether there is any significant effect of the two independent variables antidepressant drug and gender on the dependent variable rank of symptom severity. For this research study, we will use the multiple regression model for checking whether there is any significant relationship exists between the dependent variable rank of symptom severity and independent variables antidepressant drug and gender of participant. We will use ANOVA test for checking the overall significance of the relationship.

Question 4

For the given research study, we want to investigate the significant difference in the aggressive behaviour of the participants after playing normal video game and violent video game. This means we will get the bivariate data for the given research study. There would be one to one relationship between the two data sets. So, for checking the significant difference between the given two aggressive scores, we will use the paired t test for the significant difference in the population means.

Related Questions

Navigate

• Browse (All)
• Browse E
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.