Test
PRM EXAM PRACTICE
Crea una cuenta gratis para terminar este test
Guarda tu progreso y consulta tu resultado final.
¿Ya tienes cuenta? Inicia sesión
Preguntas del test (64)
1An ad hoc inference is:
- AAs reliable as a statistical inference
- BA ‘gut’ or intuitive inference based on recent past experience
- CA scientific judgement
- DBased on scientifically captured data
2Sample statistics differ from population parameters as:
- ASample statistics are represented by Greek letters
- BPopulation parameters refer to a subset from the sample
- CSample statistics are represented by Roman letters
- DSample statistics cannot be used in statistical formulae
3Cohen’s d suggests that:
- A.8 is a small effect
- B.2 is a small effect
- Cscores >1 are not possible
- D.5 is a large effect
4A correlation coefficient is:
- Adetermines is a causes b
- BUnable to determine the strength of the relationship
- Cbounded between -1 and 1
- DUsed only to determine the direction of relationship between variables
5The assumptions that need to be satisfied for Pearson correlation are
- ABivariate normality and independence
- BThat the data comes from a valid and reliable sample
- CBivariate normality and heteroscedascity
- DHeteroscedascity and independence
6Outliers:
- Acan increase, reduce or have no substantive impact on the strength of a relationship
- Bincrease the strength of a correlation
- Creduce the strength of a correlation
- DHave no impact on the relationship between variables
7The p value associated with an r value:
- ATells us if the strength of the relationship is strong
- BTells us if the direction of the relationship is statistically significant
- CTells us if the p value and r value are related
- DTells us if the r value significantly differs from the null hypothesis
8The ρ symbol refers to:
- AThe sample correlation value
- BThe probability of the r value being significant
- CThe population r*
- DThe direction of the relationship between variables
9A Type I error refers to:
- ACases when the null hypothesis is incorrectly rejected
- BMaking an error in calculating the statistic (No, this is not correct, though this may cause a Type 1 error)
- CCases when the null hypothesis is correctly rejected
- DCases when the alternate hypothesis is correctly accepted
10Simple linear regression:
- ADoes not give an indication of the strength of a relationship between variables
- BCan involve measuring the association of several independent variables with the dependent variable
- CGives an indication of the strength of the relationship between variables and is used for assessing straight line associations
- DIs best used for estimating curvilinear relationships
11The advantage of reporting standardised statistics is:
- AThat it tells us if the variables are significantly related to each other
- BThat a person unfamiliar with the variable properties can still interpret the effect size
- CThat it explains less variance than the unstandardized form
- DThat it explains more variance than the unstandardized form
12X is the independent variable and Y is the ..
- Amediator
- Bdependent variable
- Cpredictor variable
- Dslope variable
13A scatterplot….
- Acan be used to assess if two variables are significantly correlated
- Bis used as a last resort to identify outliers in the data set
- Cuses a formula to calculate the p value
- Dcan be used to identify heterogeneous subsamples and outliers in the data set
14Removing outliers from the dataset…
- Amay reduce the ability to report a significant finding as the sample size will decrease
- Breduces the strength of the association between variables
- Cincreases the strength of the association between variables
- Dis inappropriate as you are altering the dataset
15‘Fanning’ of the data points to one side of the scatterplot may indicate:
- Anon-normality
- Bheterogeneity of variance
- Clinearity
- Dhomogeneity of variance
16Multiple regression involves:
- AThe use of several simple regressions to produce a mean
- BThe assessment of the relationship of multiple predictors on multiple dependent variables
- CThe assessment of the relationship of multiple predictors on Y
- DThe assessment of multiple Y’s on X
17Multicollinearity is…..
- Aan assumption that needs to be satisfied to proceed with a multiple regression
- Ba necessary element to proceed with a multiple regression
- Ca problem as highly correlated independent variables violates the assumption for multiple regression
- Da problem as highly correlated dependent variables violates the assumption for multiple regression
18A multiple regression may not be statistically significant due to:
- Athe R² being too large
- Bthe R² being too small
- Cthe p value being too small
- Dthe sample being too large
19If we assessed if the use of various psychological techniques were related with improved sporting performance and the finding was that the set of 3 techniques predicted performance R² = .41, F(3, 25), p = .04. We could state:
- AThat the finding was non – significant
- BThat there were 25 independent variables
- CThat there were 29 scores in the data set
- DThat there were 4 independent variables
20You are more likely to attain a higher R² when:
- AYour independent variables are highly correlated
- BYour independent variables capture substantial unique components of the dependent variable
- CYour set of independent variables share substantial variance with each other
- DYour dependent variable is not related with the independent variables
21If we are looking for evidence to support the hypothesis that 3 psychological techniques predict sporting performance, we would hope that:
- AAll three independent variables are uniquely related to the dependent variable
- BThe three predictors are highly correlated with each other
- CThe dependent variable is not too highly correlated with the independent variables
- DThe tolerance statistic of several independent variables is < .10
22R² is derived from…
- AThe combination of the shared variances between all Y’s with the X
- BThe combination of unique contributions from each predictor
- CThe combination of the shared variances between all X’s with the Y
- DThe combination of unique and shared contributions of the X’s with Y
23Three types of statistical regression are:
- AHierarchical, Forward, Backward
- BReverse, Forward, Statistical
- CBackward, Stepwise, Forward
- DStairwise, Forward, Reverse
24When performing a Forward regression compared to a standard multiple regression using the same number of predictors….
- AA larger N would be required for the Forward regression
- BThe same N would be required
- CA smaller p criterion is set
- DA smaller N would be required for the Forward regression
25You have been exposed to various forms of regression so far- including:
- ALogistic regression
- BReduction regression
- CStepwise regression
- DPoisson Regression
26Hierarchical regression is not useful for:
- ARemoving the effects of nuisance variables
- BHypothesis testing
- CReducing predictors in a data set
- DCovariance analysis
27If the R²change is not significant in a hierarchical regression we can say that:
- ANone of the variables were related to the DV
- BThe group of predictors in this step did not improve the model fit
- CThe predictors shared too much variance
- DThe N was too small
28It is NOT possible to use a measure as an independent variable if the format is:
- ALa Trobe University, Murdoch University, Melbourne University
- B< 20 years old, 20-29 years, 30-39 years
- CHeight in millimetres
- DVoting Yes/No
29A ‘dummy variable’ is:
- AA variable created to recode a polytomous predictor
- BA polytomous predictor
- CA variable that cannot be used in regression
- DA scalable polytomous variable
30Heterogeneity of subsamples can be best assessed by:
- ABy assessing if slopes ‘cross-over’ each other
- BStudying separate residual plots for each subgroup
- CBy using a hierarchical regression and assessing if the intercepts differ between subsamples
- DBy using a hierarchical regression and assessing if the slopes for each subsample differ from parallel
31The ability to generalise our findings from an experimental design is governed by:
- Athe sample of participants in the study
- Bhow the variables have been operationalised
- Cthe sample of participants in the study AND a cross sectional design
- Dthe sample of participants in the study AND how the variables have been operationalised
32Internal validity may be improved by:
- Areducing measurement error AND
using more homogenous samples AND
using a control group - Breducing measurement error
- Cusing a control group
- Dusing more homogenous samples
- Areducing measurement error AND
33When we ‘test’ participants in an experimental setting, internal validity may be compromised due to:
- Anon-applicability to real-world settings
- Bsmall sample
- Cnon-applicability to real-world settings AND small sample
- DHawthorne effects
34What makes an experiment an experiment?
- Arandom allocation of participants to groups
- Brandom sampling AND controlled extraneous variables
- Ccontrolled extraneous variables
- Drandom sampling
35The Solomon group design is useful as it allows for an assessment of:
- Athe effects of post-testing
- Bthe effects of pre-testing
- Cdifference scores between groups AND the effects of pre-testing
- Ddifference scores between groups
36Random allocation to groups helps control for:
- Aindividual differences
- Bparticipant mortality
- Cexperimenter bias
- Dsmall effect sizes
37Multiple t-tests are not recommended because:
- Athey increase the Type 1 error rate
- Bthey inflate the experiment wise error rate AND
they increase the family-wise error rate AND
they increase the Type 1 error rate - Cthey increase the family-wise error rate
- Dthey inflate the experiment wise error rate
38 The obtained F value of a study that uses a one-way ANOVA to test for differences between 4 means is 2.30, this means that:
- Aas it is over 1, you would need to conduct post-hoc testing
- Bthe result is significant
- Cthe groups differ
- Dthe between group variance was greater than the within group variance, but we don’t know if it is significantly greater
39A one –way ANOVA produces an eta square effect size of .17. This means:
- Athe effect is likely due to chance
- Bthe result will be significant as the effect is large
- Cthe effect is medium strength
- Dthe effect size is large
40When conducting a one-way independent ANOVA with three levels on the independent variable, an F-ratio that is large enough to be statistically significant tells us:
- Athat there is a significant three-way interaction
- Bthat the model fitted to the data accounts for less variation than extraneous factors, but it doesn’t tell us where the differences between groups lie
- Cthat all of the differences between means are statistically significant
- Dthat one or more of the differences between means is statistically significant but not where the differences between groups lie
41The critical difference score attained after a post-hoc Tukey test tells us:
- Athe difference between means that needs to be matched or exceeded to accept the alternate hypothesis
- Bthe effect size of the comparison
- Cnothing until it is compared to the Q table
- Dthe maximum distance apart the means can be
42The Student Neuman-Keuls test is used in preference to the Tukey when:
- Awe have 3 or more group means
- Bwe have 3 group means
- Cwe have small effect sizes
- Dwhen the Tukey test is too conservative
43Post-hoc setwise comparisons:
- Ause Tukey or Student-Neuman-Kuels tests to calculate critical difference scores
- Bare used after you have found none of the pairwise comparisons are significant
- Cdiffer to pairwise comparisons because they are less conservative
- DWhen you wish to compare several group means against another mean(s)
44In order to ensure that our contrasts are linear and orthogonal:
- Athe product of each column wise contrast (e.g., contrast 1 x contrast 2) sums to 0
- Bwe need to have no more than K-1 contrasts AND
the product of each column wise contrast (e.g., contrast 1 x contrast 2) sums to 0 AND
the sum of all coefficients in a single contrast sum to 0 - Cthe sum of all coefficients in a single contrast sum to 0
- Dwe need to have no more than K-1 contrasts
45Trend analysis is useful:
- Aif you want to test a theory that the positive relationship between the number of psychology treatment sessions and improved self-esteem are not linear
- Bif you want to compare if a course of anti-inflammatory treatment reduces depressive symptoms between very high, high, medium and low and very low socio-economic groups
- Cif you want to compare if a course of anti-inflammatory treatment reduces depressive symptoms between very high, high, medium and low and very low socio-economic groups AND
if you want to test a theory that the positive relationship between the number of psychology treatment sessions and improved self-esteem are not linear - Dif you want to determine if differences in IQ between autistic, Downes syndrome or Fragile X populations follow a quadratic trend
46Planned comparisons:
- AAre used after you have a significant ANOVA finding
- BUse the Scheffe test to control for Type 1 error
- CUse the Tukey test to control for Type 1 error
- DUse the Bonferroni or Holm’s methods to distribute the .05 alpha level
47What is the correct list of these tests from most to least powerful?
- AScheffe, Tukey, SNK
- Bt-test, Tukey, SNK
- CBonferroni, Holms, SNK, Tukey
- DHolms, SNK, Tukey, Scheffe
48Two-way ANOVA differs from One-way ANOVA:
- ABecause you can calculate main effects and an interaction effect
- BBecause you can analyse the data two ways
- CBecause Two-way ANOVA is for two groups
- DBecause one uses a F test and the other does not
49A 4 x 4 ANOVA would be:
- AAn example of a two way ANOVA
- BAn example of a 8 factor ANOVA
- CAn example of a sixteen way ANOVA
- Dthe product of each column wise contrast (e.g., contrast 1 x contrast 2) sums to 0
50An interaction effect is achieved when:
- AWhen the partial eta square is > .14
- BWhen the plot of the group means shows that the lines crossover
- CWhen the main effects are not significant
- DThe p value for the F test is significant
51If our 2 x 2 ANOVA shows we have a significant interaction effect:
- AWe should report the main effects in preference
- BWe would need to conduct further analyses to understand how factors impacted upon each other
- CThere is no need to do additional analyses to understand the results
- DThe homogeneity of the test is questionable
52In a 2 x 8 ANOVA, a simple main effects analysis:
- AWould involve 10 simple main effects analyses
- BWould involve 2 main effects analyses
- CWould involve 16 main effects analyses
- DIs only used when you have more than 2 levels in a factor
53Post-hoc tests of a simple main effects analysis:
- AAre used after a significant main effect when there are more than 2 levels in the factor
- BAre generally considered redundant
- CWould be used if you had a non-significant simple main effect
- DAre not necessary when you have 3 or more levels in your factor
54Tau is a statistically contrived indication of:
- AThe normal fluctuations we see in an individual over time
- BAn assessment of individual and systematic variation
- CHow people respond to changes in the environment
- DBetween-subject variation
55The main reason we use ABBA designs over ABAB designs is:
- AABAB designs are good for practice but not fatigue effects
- BABAB designs are more likely to falsely suggest treatment effects than ABBA designs
- CABBA designs are shorter duration
- DABBA were a great band
56Incomplete within-subjects counterbalancing designs may be used in preference to a complete-within-subject counterbalancing design because:
- AThe researcher wants to use ABBA counterbalancing
- BThe researcher wants to achieve higher scientific rigor
- CThe researcher wants to allocate more participants to each sequence
- DThe researcher want to test all subjects with the same controlled sequence
57Within subject ANOVA separates the within subject variability from the between treatment variability:
- ABecause we only want to assess the within group effect
- BTo allow a comparison of between and within sources variation
- CBecause this is a source of error that when removed decreases the chance of a statistically significant result
- DBecause compound symmetry needs to be violated
58When Mauchly’s test of sphericity is non-significant:
- AWe know that there are no differences between groups
- BWe have violated the assumption of compound symmetry
- CCompound symmetry assumption is not violated but sphericity is violated
- DThere is not a significant interaction between individuals and treatment and we can have confidence the F test is valid
59When choosing a post-hoc test after a within group ANOVA, we DON’T need to consider:
- Aour sample size
- Bthe error rate
- Chow many means are being compared
- Dthe error rates
60An ANCOVA is most similar conceptually, to:
- AHierarchical Regression
- BForward Regression
- CStepwise Regression
- DLogistic Regression
61An appropriate covariate reduces the error in an ANCOVA model and this:
- AIncreases the Type 1 error rate
- Bmakes finding a significant model less likely
- CIs necessary to satisfy the assumption of normality
- Dmakes finding a significant model more likely
62When using an ANCOVA, the homogeniety of slopes:
- Ais less important than heterogeneity of slopes
- Bis not a problem if you do not block the covariate
- Care assessed using a tw-way ANOVA
- Dis an assumption that needs to be satisfied before proceeding
63The difference between a A One way ANOVA and a post hoc test is:
- APost-hoc tests are more powerful than ANOVA
- BThe ANOVA is exploratory and the post hoc tests are based on theory
- CThe ANOVA tests if the within variance is significantly less than the between group variance
- DA one way ANOVA has several factors and the the post hoc-test has only one
64The Mauchly test is an assessment of:
- ASphericity
- BHow much I hate stats!
- CHomogeneity of slopes
- DNormality
Información
- Descripción
- Test sobre PRM EXAM PRACTICE, creado por Katie Garth el 30/05/2018.
- Autor
- Katie Garth
- Idioma
- English
- Visitas
- 0
- Valoración
- Aún sin valoraciones Valora este test