Exam Qs Flashcards
In the context of ANOVA, the term ‘MS’ refers to
Variability
What do you need to know to calculate the MS in an ANOVA
Sum of squares
df
How many null hypotheses are typically associated with a 2-way factorial ANOVA?
3
What is the main advantage of undertaking an experiment with a repeated measures design over an independent-groups design
You don’t need to include as many participants
A researcher conducts a study that has 3 factors.
Each factor has 2 levels.
All participants take part in all conditions.
What type of ANOVA would be appropriate?
2x2x2 repeated-measures ANOVA
When should a mixed factorial ANOVA be avoided?
When all participants undertake all conditions.
When there are only 2 groups of participants undertaking a single condition
When there is only 1 group of participants undertaking only 2 conditions
Which of these tests offers the highest level of protection against type 1 errors?
Tukey’s HSD
Scheffe’s
Fisher’s LSD
Bonferroni
Tukey’s HSD
Which of these corrections is related to sphericity?
LSD
Hugh-Feldt
Bonferroni
Hugh-Feldt
Is sample size an ethical issue?
Yes
Which of these tests is most prone to type-2 errors?
Tukey’s HSD
Scheffe’s
Fishers LSD
Bonferroni
Bonferroni
Which tests have greater statistical power?
Parametric or non-parametric
Parametric tests
What type of test should a researcher use to analyse nominal data?
Chi-squared
Which of the following is an example of a true experimental design?
Casual comparison
Static group comparison
Convenience sampling
RCT
RCT
Which of these is an example of a quasi-experimental design?
RCT
Solomon 4group design
Static group comparison
Pre-experimental design
Static group comparison
Which of the following tests in the non-parametric equivalent of the independent samples t-test?
Wilcoxon signed-rank test
Friedmans ANOVA
Mann-Whitney U test
Spearman’s Rank order correlation
Mann whitney u test
Which declaration is a statement of ethical principles for medical research involving human subjects, including research on identifiable human material and data?
The Helsinki Declaration
In the context of ANOVA, variability is the
Mean square
How is mean square calculated?
Sum of squares / df
When conducting post-hoc comparisons in the 1-way independent ANOVA, what procedure is most likely to yield statistically significant differences between group means?
Tukeys HSD
In multiple regression case wise diagnostics, if you had a normal sample, 95% of cases would have standardised residuals with what range
-2 to 2
A study has 3 groups each with 8 participants.
If the SSm for this study is 38.26, what would the SSbetween be?
38.26 x 8 = 306.08
In the formula for Chi-square what does O mean
Observed frequency
How many null hypothesis could be associated with a 3x2x2 repeated measures factorial ANOVA?
3
Which value is most relevant in an experimental context, population variance or sample variance?
Sample variance
Which value tells us the total dispersion of values in the data set from the mean?
Sum of squared errors
Why might the use of a 2 group pre and post test study give you an advantage over one group pre and post test study
By measuring twice, the researcher can establish an effect within 1 group that may result from exposure to a manipulation.
Define inter-rater reliability
Measures the degree of agreement between different people observing/assessing the same thing
Define split-half reliability
A test for a single knowledge area is split into 2 parts + then both parts given to 1 group of students at the same time.
What does it mean if the Durbin-Watson statistic has a value close to 2?
The assumption of independence of errors has been met
How many DV and IV are in a 2-way fully repeated measure ANOVA
1 DV
2IV
What are the consequences to the degrees of freedom after applying the Greenhouse-Geisser correction factor?
Value of the Greenhouse-Geisser epsilon = 0.934.
So the Df for each of the Greenhouse-Geisser tests are 0.934 times the Df for the Sphericity Assumed tests.
When assessing whether the results of a study conducted on a sample of participants can be generalised to the population, what are we referring to?
External validity
How can you calculate how many groups took part in the study from the ANOVA SPSS output?
df between = no. of groups -1
So if df between groups = 5
Then 5 + 1
Calculation for the df interaction
(n-1)(k-1)
no. of columns -1)(no. of rows-1
