Week 5-6 psychological measurement Flashcards
What’s measurement?
Assigning a score to an individual to represent certain characteristic
What’s psychometrics?
Psychological measurement, requires systematic procedures to establish reliability and validity
why are operational definitions important?
Important to understand what people are measuring and mainupating important also so we can replicate their studys
What is the most common measurement in psychology, most common for describing and most common in sciences? (NOIR)
Common in psych: ordinal
common in describing: nominal
common in sciences: interval/ratio
What is every observed score made up of? (2 things) true and random..
- true mean
- random error
what’s the line of best fit?
Line that gets closest to every data point without moving far away from other data points
What are the three types of reliability’s? and explain (Test.., Inter.. (2 things go in this one Cron, Split 1/2), Inter ra.. (1 thing goes in here, frat name almost)
Test re-test reliability - measure twice to see difference/match of scores
Internal consistency reliability- 2 types fall under: 1) Cronbacks Alpha-2 sets of questions to see correlation as a group2) Split half reliability-split set of q’s in two halves to see correlation
Inter rter reliability: use correlation to see whether 2 ppl will score high or low
1 thing falls under: Cohens Kapa
What does R2 and 1-R2 represent? (variance, coefficient)
R2 is variance accounted for and
coefficient of determination
1-R2 is variance unaccounted for and coefficient of alienation
What are the 3 types of validity’s and explain them? (ex, in and other is easy to remmeber)
validity- how well score represents what it is supposed to
External validity- how well does study model population you want to make statements about
Internal validity- degree of confidence causal relation isn’t influenced by other factors
6 ways to assess validity (F, C,P,C,C,D)and meanings
Face validity- Does the construct look the way it’s supposed to
Construct validity- is the construct being measured accurate?
Concurrent validity- compare measure against different valid measure
Predictive validity- how well can measure predict behaviour
Convergent validity- do other methods have similar measures
Divergent validity- ensures extremely different measures are extremely different
What is statistical significance
?
for researcher to say look this paper is right and you should believe it as it’s statistically significant and is related to probability of events happening by chance significance doesn’t make it important though
What’s effect size? (3 different things) (hypotheses discrepancy, degree to which.., impact)
- discrepancy between null and research hypothesis
- Degree in which null hypothesis is false
- impact of Iv on Dv
What are the three sizes of correlation
small- 0.1
medium-0.3
large-0.5
6 Threats to external validity and explain (P,R,P/E,T,Levels of.., M AND Pro..
Person- who’s statements about ? do they apply elsewhere?
Researcher-can you obtain same results by different researcher?
Environment/place- would you obtain same results in different environment?
Levels of treatment- does it match real world?
Measurements,procedures- would others get same results
What are the threats to internal validity and explain them
- factors creating doubts IV caused results
- History- external events that impact subject
- maturation- changes within a participant
- Instrumentation- rater who measures criteria must be consistent
why do we use generalizability?
To see how much weight we should be putting in a paper
Strengths and weakness of independent group designs (2 for each) effects and lower.. Dis: matches,2 times the ..
Adv
No order effects
lower drop out rate
Disadv
Groups may not be well matched , need double participants
Strengths and weaknesses of repeated measure designs (2 for advantages and 1 for disadvantages) Adv: perfectly.., need less Disadv: may experience
Perfectly matched groups
need less participants
Disadv
may experience order effects like boredom or fatigue
What are order effects and how we can reduce them? (2 ways to reduce: I and C)
Way participant responds differently in result of the order of research (getting bored of it ) 2 ways to control for this are
reducing interval times between treatments/sections
Counterbalancing - splitting groups in 2 so goes quicker through treatments/sections
What are the three types of control conditions and explain (Trea and c, P and C, Wait.. )
Treatment and control group- one group receives treatment control doesn’t receive
Placebo and control condition- placebo group gets fake sugar pill control group gets real medication
Waitlist control condition- used to combat change over time (maturation)
Why do we use pre and post tests? (eliminates.. differences) and when is the pre test given?
As it allows us to look at similarity’s or differences between the two scores and eliminates alternative causes not caused by the IV
Before experimental manipulation
What are the big 5 validity’s? (V, S,E,I,C)
Validity
Statistical validity
External validity
Internal validity
Conclusion validity
