Quiz
307 Q2 Meta-Analysis
Dieses Quiz enthält Fragetypen, die sich im gewohnten Player öffnen.
Fragen in diesem Quiz (26)
1Meta-analysis is a good method for combining the results from multiple studies?
- ARichtig
- BFalsch
2Who was the first person to use meta-analysis?
- AKarl Popper
- BCarl Pearson
- CGene Glass
- DAaron Beck
3Meta-analysis is not a statisitical method
- ARichtig
- BFalsch
4Meta-analysis combines the results from which types of studies?
- AQuantitative
- BQualitative
- CQualitative and Quantitative
- DIt is not used to combine studies
5Meta-analysis helps to overcome which problem with small sample sizes?
- APower
- BReliability
- CValidity
- DCorrelation
6Select the three categories of analysis as defined by Glass
- APrimary analysis
- BSecondary analysis
- CMeta-analysis
- DOriginal analysis
- EStatistical analysis
- FRe-analysis
7Glass categorised three levels of analysis
[blank_start]Primary analysis[blank_end]: The original analysis from the study.
[blank_start]Secondary analysis[blank_end]: Re-analysis of original data from a study using different statistical techniques.
[blank_start]Meta-analysis[blank_end]: Statistical analysis of results from a large number of individual research studies so as to integrate their findings.- APrimary analysis
- BSecondary analysis
- CMeta-analysis
8Effect size appears to overestimate the true score slightly
- ARichtig
- BFalsch
9Overestimation of the effect size becomes trivial when sample sizes are greater than...
- A10
- B15
- C20
- D25
- E30
10Effect sizes are used in their original form in meta-analysis
- ARichtig
- BFalsch
11[blank_start]Fixed effect model[blank_end]: Residual variance is low, effect is similar across the population due to homogeneity. Effect sizes are weighted according to sample size. Should only be used with [blank_start]homogeneous[blank_end] populations.
[blank_start]Random effect model[blank_end]: Residual variance is high, effect differs across the population due to heterogeneity. Effect sizes are weighted according to variance and sample size. Must be used with [blank_start]heterogeneous[blank_end] populations, can be used with homogeneous (and is the preferred).- AFixed effect model
- Bhomogeneous
- CRandom effect model
- Dheterogeneous
12The fixed effect model can be used with both homogeneous and heterogeneous populations.
- ARichtig
- BFalsch
13The random effects model can be used with both homogeneous and heterogeneous populations.
- ARichtig
- BFalsch
14Effect size is also know as
- Ad
- Bp
- Co
- De
15Cohen's conventions are
- ASmall - .1; medium = .3; large = .5
- BSmall - .3; medium = .5; large = .7
16The strength of an effect size (d) is compared to which conventions?
- ACohen's Small - .1; medium = .3; large = .5
- BPearson's Small - .3; medium = .5; large = .7
17The strength of an effect size is compared to Cohen's conventions, however, Glass stated that it does depend on the situation.
- ARichtig
- BFalsch
18When residuals vary consistently across a population the population can be considered
- AHeterogeneous
- BHomogeneous
19When residuals vary randomly across a population the population can be considered
- AHeterogeneous
- BHomogeneous
20The fixed effect model is the preffered model for weighting with meta-analysis
- ARichtig
- BFalsch
21It is desirable for a sensitivity analysis to be included in a meta-analysis
- ARichtig
- BFalsch
22It is desirable for a meta-analysis to include a sensitivity analysis to determine the robustness of the results.
Performed by analysing the data with some studies [blank_start]removed[blank_end] from the analysis. Done by testing different ways the dataset can be ‘sliced’. If the results [blank_start]change dramatically[blank_end] with a study or studies removed, we might question the credibility of the conclusion.- Aremoved
- Badded
- Cchange dramatically
- Ddon't change
23A strength of a meta-analysis is publication bias, where journals are biased against including null result articles.
- ARichtig
- BFalsch
24Publication bias can lead to p-hacking (where results are tweaked to improve the significance value)
- ARichtig
- BFalsch
25Meta-analysis can be prone to selection bias meaning researchers only select certain studies to produce a specific result.
- ARichtig
- BFalsch
26Even though meta-analysis has weaknesses, they are unlikely to overestimate the true effects.
- ARichtig
- BFalsch
Informationen
- Beschreibung
- Research Methods Quiz am 307 Q2 Meta-Analysis, erstellt von Ryan Bentham am 19/09/2018.
- Autor
- Ryan Bentham
- Sprache
- English
- Aufrufe
- 1
- Schlagwörter
- Bewertung
- Noch keine Bewertungen Bewerte dieses Quiz