Week 5 - Statistical models (ANOVA and non-parametric alternatives) Flashcards

1
Q

A study has been run on eating behaviours and body image. Participants were n=1000 1st year university students in Melbourne, Australia. Eligibility criteria included: 18-40 years of age, English as 1st language, and no history of disordered eating. It was reported that those with more variable eating behaviours across a week had significantly poorer body image, with a medium effect size. What can we say about this effect?
a. It is a strong and impressive effect
b. It appears in biased sample (young adult, highly educated, with English language background) and may not generalise to more diverse samples
c. It appears in a relatively unbiased sample and will likely generalise to the general population.
d. It appears in biased sample (young adult, highly educated, with English language background) but will likely generalise given the medium effect size

A

b. It appears in biased sample (young adult, highly educated, with English language background) and may not generalise to more diverse samples

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

You are writing a psychological research report. What is not a Method sub-section heading?
a. Measures
b. Conclusion
c. Participants
d. Statistical Approach

A

b. Conclusion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

A study on gambling behaviours was carried out in a laboratory setting. They reported that in a sample of n=80 middle-aged adults, that those with poorer impulse control took more risks on the gambling task, a statistically significant effect with a small effect size. What would be a correct interpretation of this finding?
a. It’s important as it’s statistically significant
b. There is a relationship between poor impulse control and riskier gambling in a laboratory setting, but this is only a small effect outside the lab
c. There is a small relationship between poor impulse control and riskier gambling in a laboratory setting, but we do not know if this generalises outside the lab
d. We can’t say much given the small effect size

A

c. There is a small relationship between poor impulse control and riskier gambling in a laboratory setting, but we do not know if this generalises outside the lab

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

We find a statistically significant correlation, with a medium effect size, between self-efficacy and academic achievement. Data were collected using a cross-sectional approach in a large sample of university students (n=5000). How can we interpret this effect?
a. Self-efficacy causes changes in academic achievement.
b. The relationship would be more important if there was a strong effect size.
c. Academic achievement causes changes in self-efficacy.
d. Self-efficacy and academic achievement are related.

A

d. Self-efficacy and academic achievement are related.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

A researcher is interviewed on ABC News. They say that they have developed a new psychological therapy for depression. They found a statistically significant difference between their new psychological therapy and standard care, in a group of adults with a new diagnosis of depression. It prompts you to access and read the peer-reviewed paper where this trial is reported. What statistic should you look for to understand how impactful this new psychological therapy is?
a.t-statistic (between-groups)
b. p value
c. Alpha value
d. Effect size

A

d. Effect size

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

We run two independent samples t-tests, comparing different types of reading interventions for students in years 1-3. For each, the DV is reading fluency. The first t-test has a p value of .003, and the second has a p value of .047. What can we say about the strength of effects?
a. The first has a strong effect and the second has a weak effect.
b. Nothing, as p values are not measures of effect size.
c. The first t-test has a stronger effect than the second.
d. The second t-test has a stronger effect than the first.

A

b. Nothing, as p values are not measures of effect size.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Two researchers run the same study design in two different locations: Streaky Bay and Port Pirie. The study involves measuring fruit and vegetable consumption and its associations with cognitive function in older people; they use the same measures and protocol. The researcher in Streaky Bay recruits 62 participants and the researcher in Port Pirie recruits 201 participants. Each researcher analyses their own sample. What would be correct to say?
a. The Port Pirie sample will have more statistical power than the Streaky Bay sample.
b. They have the sample statistical power, as they use the same measures.
c. They have the sample statistical power, as they use the same protocol.
d. The Streaky Bay sample will have more statistical power than the Port Pirie sample.

A

a. The Port Pirie sample will have more statistical power than the Streaky Bay sample.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

A researcher is aiming to investigate the experiences of refugees from Syria, Venezuela and Afghanistan living in Australia. The researcher wants to capture a diverse sample of experiences and backgrounds. They employ a qualitative approach, conducting interviews with each participant. The researcher can only speak English. What will be a critical consideration?
a. That all participants use English, so the researcher can understand
b. Ensuring translators are present at interviews, if required, so participants can voice their opinions in the language they feel most comfortable using
c. That all advertisements are in English
d. That in advertisements, it is stated that participants must be proficient in English

A

b. Ensuring translators are present at interviews, if required, so participants can voice their opinions in the language they feel most comfortable using

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

A good experimental research design does what?
a. Minimises plausible alternative explanations
b. Guarantees a p value <.05
c. Includes multiple measures of each IV and DV
d. Sets a conservative alpha value

A

a. Minimises plausible alternative explanations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

You are writing a psychological research report. What would be an inappropriate reference source?
a. Wikipedia
b. Journal of Applied Psychology
c. Psychology and Aging
d. Psychological Bulletin

A

a. Wikipedia

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

in relation to the hypothesis, ‘Intellectually disabled adolescents given different types of skills training programs (Group 1=Social Skills, Group 2=Daily Living Skills, Group 3=Both Social Skills and Daily Living Skills) will differ in their level of Assertiveness (measured on the Goldberg Assertiveness Scale).’
What is the inferential statistical test of choice?
a. Oneway ANOVA
b. Independent samples t-test
c. Pearson Correlation
d. Paired samples t-test

A

b. Independent samples t-test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Homoscedasticity is Most commonly tested using ____ test ?
a. t-test
b. Wilcoxon test
c. Kruskall-Wallis test
d. Levene’s test

A

d. Levene’s test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

If p<.05, you have _____ homogeneity of variance.
a. not violated
b. violated
c. confirmed
d. assumed

A

b. violated

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Which test is non-parametric?
a. Pearson correlation
b. Wilcoxon test
c. T-test
d. Anova

A

b. Wilcoxon test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Select the option that best describes whether the data is parametric or nonparametric, and what the most appropriate measures of central tendency and dispersion would be: Percentage score on a reading test designed for six- to seven-year-olds (that was given to a group of 10- to 15-year-olds).
a. Parametric, median, range
b. Nonparametric, mode, variation ratio
c. Parametric, mean, standard deviation
d. Nonparametric, mean, standard deviation

A

a. Parametric, median, range

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Select the option that best describes whether the data is parametric or nonparametric, and what the most appropriate measures of central tendency and dispersion would be: Age in months for a primary school year 2 class.
a. Parametric, median, range
b. Nonparametric, mode, variation ratio
c. Parametric, mean, standard deviation
d. Nonparametric, mean, standard deviation

A

c. Parametric, mean, standard deviation

17
Q

What represents effect size for an Anova?
a. d
b. η2
c. F
d. p

A

b. η2

18
Q

Your DV is continuous and non-normally distributed. Your IV has two levels (group 1 v group 2). What test should you use to compare the DV between the groups?

a. T-test
b. Wilcoxon test
c. ANOVA
d. Spearman correlation

A

Wilcoxon test

19
Q

You are running a one way between groups ANOVA. What is not an assumption check you need to run?

a. Outliers on DV (normality assumption)
b. Levene’s (homogeneity of variances)
c. Logistic regression (to test bivariate outcomes)
d. Shapiro Wilk on DV (normality assumption)

A

Logistic regression (to test bivariate outcomes)

20
Q

You run an ANOVA with six groups (i.e. six levels on your IV). You get a significant F and undertake post-hoc testing. How many comparisons will the post-hoc tests include?

a. 62
b. 15
c. 6
d. 5

A

15

21
Q

You have one continuous variable and one categorical variable (with three levels). What would be an appropriate way to visualise the relationship between these two variables?

a. Scatterplot without correlation line
b. Pie chart
c. Scatterplot with correlation line
d. Violin plot

A

Violin plot

22
Q

You are a psychologist studying the effect of different types of therapy (Cognitive-Behavioral Therapy, Psychoanalytic Therapy, and Humanistic Therapy) on reducing anxiety levels. You randomly assign participants to one of these three therapy groups and then compare their post-therapy anxiety scores using __________ to see if there is a significant difference among the therapies.

A

ANOVA

23
Q

You are researching whether there’s a relationship between gender (male or female) and preference for a particular genre of movies (action, romance, or comedy). You collect data from a random sample of people and use a(n) ________________ to determine if there is a significant association between gender and movie genre preference.

A

chi-squared test

24
Q

You want to compare the reaction times of two groups of participants: one group receives a new drug, and the other group receives a placebo. You use the _________________ test to determine if there is a significant difference in reaction times between the two groups.

A

Mann-Whitney U Test

25
Q

You are testing the impact of a mindfulness meditation program on stress levels. Participants’ stress levels are measured before and after the program. You use the _______________ test to assess whether there is a significant change in stress levels after participating in the program.

A

Wilcoxon signed-rank

26
Q

You are studying the effect of different teaching methods (lecture, online modules, and hands-on activities) on student satisfaction. You collect data on student satisfaction scores and use the ______________ test to determine if there is a significant difference in satisfaction levels among the teaching methods.

A

Kruskal-Wallis

27
Q

You are investigating the relationship between the amount of exercise people engage in per week and their self-reported happiness levels. You collect data from a group of participants and use ____________________ to measure the strength and direction of this relationship when the data is not normally distributed.

A

Spearman’s rank correlation

28
Q
A