Karteikarten
four
Karten in diesem Stapel (27)
- 1what do you need when you make a new questionnairevalidation
- 2what four things need to be considered when making a new questionnaire?1. item pool
2. pilot testing
3. reliability test
4. validity checks - 3what is the item pool and what would we do?what questions might we ask?? create an item pool of maybe 10 questions each
- 4what is pilot testing and what would we do?do our questions get at what we are asking? collect data twice from the same group and analyse it
- 5what is the reliability test and what would we do?e.g. test-retest, item analysis
can we rely on the scale? from one time to the next? do the questions appear to point in the same direction?
correlated time 1 and time 2 retest data, item analysis - 6what is the validity check and what would we do?e.g. discriminative validity 'known groups'
does our scale measure what we want it to measure? can we use it meaningfully?
compare two groups that should score differently on the questionnaire if it works - 7what can you learn from factor analysisunderlying structure of the construct we are measuring
- 8what items does factor analysis work onitems which have correlated answers
- 9how does factor analysis interpret databy combining our understanding of the phenomenon with numeric output of the analysis
- 10what does factor analysis show us about peopleshows how people appear to conceptualise certain items on the scale
- 11what are three words involved in interpreting factor analysescommunalities, variance explained, rotated component matrix
- 12what are communalitieshigh (0.7 - 0.9) which means item contributes to overall solution
- 13what is variance explainedthe more variance a factor (component) explains, the more of the variation in responses across the whole questionnaire it can account for
- 14what is rotated component matrixtidied up components, varimax finds factors with the least amount of overlap
- 15what are three terms for intepreting factor loadingshigh factor loadings, (high) negative FL, low FL
- 16what is the cut off for high factor loadings> .5
- 17what is the cut off for a low factor loading< .5
- 18what are high factor loadingsitem belongs to a factor and must take meaning of item into account when interpreting
- 19what are (high) negative factor loadingsitem belongs to the factor but the inverse of its meaning!
- 20what are low factor loadingsitem is not correlated with the factor and its meaning is not relevant to the interpretation
- 21what is another name for the extracted factorcomponent
- 22when items have been loaded, what four things do you look at?which items load, direction they load, meaning of these items (could use thematic analysis), size of loadings (larger loadings > more important)
- 23when making sense of each individual component, what four things do you look at?what do these items have in common, how can I describe the commonness to encapsulate the meaning of all the items, does the factor link to something predicted from the literature / is it new, what does it tell about peoples attitudes
- 24what are the two levels to validity of a scale?discriminative validity, construct validity
- 25what 4 things tell us if a scale is validated and suitable?is it reliable, is it valid (discriminative and construct), what does it show about construct / attitude object, how could we use it
- 26three ways to check reliabilitynumber of pilot participants, item analysis, test-retest analysis
- 27the three types of analysis in experiments such as the second practical?cronbachs alpha, factor analysis, validity check
Informationen
- Beschreibung
- PSYCHOLOGY YEAR 2 (2 rmd) Karteikarten am four, erstellt von Ki Rai am 21/03/2019.
- Erstellt von
- Ki Rai
- Aufrufe
- 2
- Bewertung