false

Study Guides
(248,398)

Canada
(121,510)

Carleton University
(4,099)

Psychology
(776)

PSYC 2002
(22)

Chris Herdman
(1)

Final

Unlock Document

Psychology

PSYC 2002

Chris Herdman

Winter

Description

Psych 2002 FINAL CHEAT SHEET
f
Proportion=p= f=frequencN=for pop.n=for sampSS=sum of squareSample Deviation= Pop. Deviation=
n
x−μ f (100) θ =SS/N ∑ x
Percentage=p(100N = Pop. Variance= Sample Mean=M= n Pop. Mean=
∑ x SS 2 SS
μ= Pop. SD=θ=√( ) Sample Variance== x= μ+z(ϑ)
N N n−1
2
∑ (x−μ) 2 ∑ x −( ∑ x )
Sum of Squared Deviation (Pop.)=SS= (def=) n (comp)Sum of Squares (sample) =
2 2 x2 SS x−μ
∑ x−M )Deff. =∑ x − ( ) (comp) Sample SD= s=√( ) z-score=
n n−1 θ
Discrete:INDIVISABLE separate categories ex.# of kids Continuous:Infinite values Nominal:Colour Ordinal:Rank by size Interval:Ordered intervals
Ratio:Interval w zero point Descriptive Stats:Describe group of raw scores Inferential Stats:Goes beyond data, can’t be manipulated Unimodal:1
Peak Bimodal:2 peaks Multimodal:2+ peaks Rectangular:0 Kurtosis:How flat/peaked a distribution is LEPTOKURTIC:Tall/Peaked MESOKURTIC:
Normal PLATYKURTIC:Broad/flat Correlational Method:No control, no manipulation, observing ex.smoking Experimental Method:1 variable
manipulated, other controlled, no confounds, cause/effect
θ θ μ
μ1−μ2=0 θM= (¿¿M) M− μ
Standard Error= √n z-test= T-test=t= sM -> if pop. Known df=
z=M−μ /¿ M
n−1 M−μ M−μ sM= s SS
Effect Size(z)=dθ Effect Size(t)=s= √n s= √ df
2 t2 2
μM=pop.mean explained variability= effect size=d=(M ¿2 s√ p
t +df
(M 1M −2) (μ1 2)
T-test for independent means=t= whether 2 means are diff. enough to be from diff. populations
S(M 1M 2)
2 2 SS +SS
p sp sp= 1 2 df =df 1df 2
sM1−M2 √(n1)+ n2 ( 11 +) (21 ) *when using this test, use to compare*
Effect size: sm: d<0.2 Md: 0.20.25
Confidence Interval=CI Determines range of values likely to contain an unknown mean. Depends on desired interval ex. 95%, 99%
μ=z θ +M
( M
Z-test T-test
H0 H 1 H 0 H 1
1. State and 1. State and 2. Set α to locate critical region. 2. Set α , determine df, find critical region on chart.
3. Collect data, preform z-test 3. Collect data, perform t-test
H 0 H 0
4. Make decision about 4. Make decision about
Statistical Power: probability a test will produce significant results if research hypothesis us true. 2 distributions can have HIGH POWER but LITTLE
OVERLAP if: 1) 2 means are very different or 2) Standard error is very small (small SD large n). Power is higher if 1) Bigger diff. between means
α
(effect) 2) More participants (lg n) 3) smaller SD. To increase power 1) Increase sample3) Use one tailed testngent
2
Explained Variability) : how much of the variability is accounted for by the treatment effect. “40% of
variability explained by effect”. T-test: Estimates pop. SD using sample SD.Type I decision error: reject
null when it is true Type II: fail to reject null when null is false.
ANOVA
1. Compute df for between-groups and within groups.
2. Calculate SS for each
3. Calculate MS for each
4. Calculate F-ratio
df
5.Find F-critical using F-table. Useand
df between
6.Compare F-obtained to F-critical.
Formulas for Indep. Measures ANOVA ^
This I for a repeated measures ANOVA.
Notation for Repeated Measures ANOVA
n # of individuals tested in aPl Participant/personal total
treatments
k Number of treatments T Sum of scores in each
group
N Number of scores G Total of all scores
* use same steps as you would use for an independent measures ANOVA
2
2 G
SStotal= ∑ x - SSwithin= ∑ SSinside eachtrea=ment1SS …2 df withi=
N
2 2
∑ T G
∑ dfwithineachtreatment SS bet.treatments − dfbet.treatments
n N
2 2
P G SS =SS −SS

More
Less
Related notes for PSYC 2002

Join OneClass

Access over 10 million pages of study

documents for 1.3 million courses.

Sign up

Join to view

Continue

Continue
OR

By registering, I agree to the
Terms
and
Privacy Policies

Already have an account?
Log in

Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.