Karteikarten
RM
Karten in diesem Stapel (43)
- 1ValidityInternal validity: ‘research
instrument to measure
what you want to measure.’
External validity: ‘Extent to which findings can
meaningfully be generalized to other settings,
situations and/or persons - 2Content and face validityContent: coverage of what u want to measure
face: At first sight - 3Criterion-related validityPredictive A causes B, possible to predict the future before B is measured
Concurrent validity A cuses B (now)
Possible to predict the future when measured at the same time - 4Construct validityConvergent: constructs that should and are related
Discriminant: constructs that should not be related but are - 5Internal validity is insufficientTheoretical model incorrect
Procedure to test the Hypotheses, not correct
other methods are not reliable - 6ReliabilityAsses whether your measurement instrument provides similar measurements when used repeatedly
Stability,Equivalence,Internal consistency - 7Internal consistencyThe degree to which multiple items measure the same construct
Cornbachs alpha - 8Nominal dataIndicate categories
Order is not meaningful
Religion - 9Ordinal dataOrder of numbers is important
Intervals are not fixed
No calculations
Clothing size - 10Interval dataOrder is important
Intervals are fixed
Basic calculations (+-)
Temperature - 11ratio dataOrder is imporatnt
intervals are fixed
all calcultions
true zero
centimeters - 12Content questionsSensitive, double-barrelled, Leading, no ambiguities, double negative
- 13Question orderFunnel approach
genarl to specific
grouping - 14Response methodsInfluence data and analysis
Open and closed questions - 15Closed questionDichotomous scale ( yes or no )
Categorical scale ( multipule choice )
Likert scale
Semantic Differential scale( multiple sacles)
Numeric scale
Graphic rating scale - 16Mutually exclusiveDon't overlap don't measure the same thing
1nothing 2 good 3 nothing - 17Collectively exhaustiveIt should cover all aspects
(plenty of other Suggestions....) - 18Closed questions: general rulesBipolar adjectives are contradictions
Balance
be specific
Sensitive topics
Don't know ( facts, not opinion ) - 19Sample vs. censusCosts
Accuracy
Speed
Availability of population elements - 20SamplesPopulations: the group you want to say something
Sampling frame: the group you can say something about
Sample: your selection of the frame - 21correct sampleAccuracy: (=absence of systematic variance)
Precision: values in the sample highly similar to values in the population ( = absence of sampling error) - 22Sample types Probability sampling ( simple)Based on random selection: everybody has the same chance
Simple: Change is >0( marbles)
impractical - 23Systematic SamplingSelects every k^th element of a population
Random starting point
Pro: Simple Similar to simple sampling
Contra: a lot of information
if the population has a trend it may be biased - 24Stratified sampleSubpopulations(=strata)
if large enough u can test differences
Pro: possibility of different methods conclusions for all strata
Contra: hard to decide, expensive - 25Cluster samplingthe population consists of clusters( cities )
- 26Non-probability samplingno random selection: different selection probabilities for the elements
subjective - 27Convenience samplingchoose whoever can be found or wants
"quick and dirty " - 28Snowball samplingRespondents identify other potential respondents for the researcher
A: efficient to find respondents in a niche
DA: similar respondents ( not representative for the population) - 29Non-responseRandon non-response: No difference between who did and who did not respond
results are correct
Larger sample
Saystematic differ systematically on imporant varibles
biased
larger better sample - 30Experimental researchDiscover causal relationships
Researchers control over experimental settings
random assignment of subjects to conditions - 31True ExperimentGroups are equal, full control
A: study something which in real life is impossible
DA: Evaluation apprehension Guinea pig effect
manipulation and setting _ less realistic - 32Field experimentReal world
heterogeneous participants
less evaluation apprehension
Manipulation
randomly assign participants
less control - 33Quasi-experimentsNatural setting
use existing groups or differences
No random assignment and/or no real manipulations less control - 34Missing dataListwise deletion
Pairwise deletion
replace by mean - 35Big three1. How many variables are involved?
2. What is the data type?
3. Asymmetric vs.symmetric - 36Related vs. unrelated sampleRelated: Each participant in each condition (small sample needed )
Unrelated: Between-subjects design - 37Repoting resultsABCD/grandma-method
A: What is the goal?
B: How did you do that?
C: What was the result?
D: Which conclusion? - 38Regression analysispositive or negative relationship
F-value(model
Rsquare:total variance expleind
t-value (Xi)
B-value of xi: positive or negative relationship - 39t-testDiffer in means
t-value
Look at means
y1>y2 or y1 - 40Aim statistical testinformation about the generalizability o the results found in study
terms of probabilities you say something about the probability that results are due to chance - 41tests: three partsTest statistic ( x^,t,F) Information about the outcome
p-value (probability of the outcome
degrees of freedom (df) - 42Pearson CorrelationA measure of the strength of a linear relationship between two interval variables
R
high value -> strong relationship - 43Spearman rank correlationhow well the relationship between two ordinal variables can be described
rho
high value->strong relationship