Easy Questions
What is the difference between independent samples t-test and paired samples t-test?
Independent samples: comparing two different/independent group means
Paired: comparing means across two different time points/conditions for the same people
What is the IV and what is the DV: A researcher gives participants a list of words to memorize. Recall is tested once in the evening and again the next morning after a full night’s sleep.
IV: time of day (evening and morning)
DV: memory recall
t(29) = -1.98, p = .09
30 and no
How do you run a paired samples t-test in SPSS?
Bonus 50 pts: check for normality (Histogram, Skewness, Kurtosis, Shapiro-Wilk Test)
Bonus 50 points: visualize your data (box plot)
Analyze > Compared means and proportions > paired samples t-test > put variables in var 1 and var 2
H₀: There is no difference in stress levels before and after a 4-week mindfulness training program.
What is the one-tailed H₁, the IV, and the DV?
H₁: Stress levels are significantly lower after the 4-week mindfulness training program compared to before.
IV: before and after training
DV: stress levels
If the mean difference is -5.3, what does that mean?
Scores from group 2 are higher than scores from group 1.
What are the effect size ranges for Cohen's d?
small = .2
medium = .5
large = .8
What statistical test should be used in this scenario and why: A group of students takes a difficult math test on two separate days. On the first day, they take it as usual. On the second day, they take the same test but are required to recite positive affirmations about their math abilities beforehand. The researcher compares their scores.
Paired samples t-test; same group of students in two different conditions
What do these statistics mean: MD = 2.1, t = 1.83, p = .075
Group 2's scores were lower than group 1 and the results were not statistically significant.
Interpret these Cohen's ds:
1) .13
2) .92
3) .37
4) .61
5) .48
6) .70
1) small
2) large
3) small-medium
4) medium
5) medium
6) medium-large
What statistical test should be used and why: A researcher wants to study how stress affects memory. They recruit 50 college students who each bring a close friend to participate in the study. The researcher randomly assigns one person from each pair to a high-stress condition and the other to a low-stress condition. Afterward, both participants take the same memory test, and their scores are compared between the high-stress and low-stress groups.
What could the paired samples t-test null and alternative hypotheses be based on these variables and statistics and should reject or fail to reject the null?
Variables: concentration and classical music
Statistics: MD = 3.7, t = 2.45, one-tailed p = .022?
H₀: No difference in concentration between classical music and no classical music.
H₁: Concentration will be better in the classical music condition compared to the no classical music condition.
Statistics: reject the null
What type of statistical test is Cohen's d?
t-test!
It measures effect size.
A researcher is investigating the relationship between sleep quality, stress, and academic performance among college students. They collect data from 100 students, measuring their sleep quality, stress levels, and GPA to determine (1) whether sleep and stress predict academic success, (2) examines whether students who sleep at least 7 hours per night have higher GPAs than those who sleep less than 7 hours, and (3) determine the effectiveness of an intervention on cognitive performance. What statistical analyses will be used for each goal?
1. multiple linear regression
2. independent samples t-test
3. paired samples t-test
APA format time!!! Everyone will be able to earn points here. Please hold while I pull up the SPSS output. Include the interpretation of the results.
A paired samples t-test was conducted to evaluate whether employees were more concerned with pay prior to or post a recession. The test was significant, t(29) = 2.83, p = .008, Cohen’s d = .52. Specifically, mean concern for pay prior to the recession (M = 5.67, SD = 1.49) was significantly greater than the mean concern for pay post the recession (M = 4.50, SD = 1.83). These results indicate that concern about salary level pre-recession when economic conditions are good is moderately greater than concern about salary level post-recession.