false

Study Guides
(248,356)

Canada
(121,501)

Simon Fraser University
(3,618)

Criminology
(360)

CRIM 320
(11)

Martin Andresen
(2)

Final

Unlock Document

Criminology

CRIM 320

Martin Andresen

Spring

Description

One-sample t test
Objective = Determine whether mean of ONE sample differs significantly from the known/estimated
population mean
o RQ: Is Vancouver’s poverty rate significantly different from national poverty rate
o Null – Vancouver rate equal to national average (H0: u= 19)
o Alt – Vancouver rate not equal to national average (1 : u ≠ 19)
One-Sample Statistics
Std.
N Mean Deviation
% 125 16 5.3437
poverty
One-Sample Test
Test Value = 19
t df Sig. (2-tailed) Mean Difference
1.556 124 .001 3
Results
1. Test of significance
Mean difference = 3
T-value = 1.556
P-value = 0.001
If (p value > .05) = Null not rejected = sample not sig different than population –
due to sampling error
If (p value ≤ .05) = Null rejected = sample significantly different than population –
greater or less than?
o Report mean – greater/lesser? (Van = 16, National = 19 – van sig less
poverty)
Effect size = mean difference / std deviation
o S (0.2) M (0.5) L (0.8)
o The amount of difference between SM and PM in terms of standard deviation units
Eg - 3/5 = 0.6 = medium
Assumptions
o Observations are independent = participants do not influence each other
o Dependent variable is normally distributed (else problematic, statistical procedure
untrustworthy, p-values inaccurate, standard deviations not useful in measuring variability) Independent samples t test
Objective = Whether Two independent groups differs significantly on dependent variable of interest
o RQ: Is there difference in average pain levels, between MIS and Traditional surgical procedure?
o Null – Pain level scores for Trad and MIS is equal in the population (H : u0 = uM)
o Alt – Pain level scores for Trad and MIS is NOT equal in the population (H : uT ≠ uM)
0
Group Statistics
procedure N Mean Std. Deviation
pain MIS 15 5.7333 1.66762
Traditional 15 7.8667 1.59762
Independent Samples Test
Levene's Test for
Equality of
Variances t-test for Equality of Means
Sig. (2- Mean Std. Error
F Sig. t df tailed) Difference Difference
pain Equal variances
.005 .943 -3.578 28 .001 -2.13333 .59628
assumed
Equal variances not -3.578 27.949 .001 -2.13333 .59628
assumed
Results
1. Levene’s Test = test for whether population variances are equal or not
F-value (0.005)
P-value (0.948) greater, null not rejected, variances are equal, first row
p value > .05 = null NOT rejected, population variances equal = FIRST ROW
p value ≤ .05 = null rejected, population variances NOT equal = SECOND ROW
2. Test for equality of means
T-value = 3.578
P-value = 0.001
If (p value > .05) = Null not rejected = pain levels same b/t two groups – sampling
error
If (p value ≤ .05) = Null rejected = pain levels differ b/t two groups
o Report mean – which lower/higher (MIS = 5.7, Trad = 7.8 -- MIS had less
pain)
Effect size = t √(N1+n2/(n1n2))
o S (0.2) M (0.5) L (0.8)
o The amount of difference in scores between 1 and 2 in terms of standard deviation units
Eg - -3.578√(15+15/(15*15)) = -1.31 = very large MIS 1.31 SD lower pain than Trad
Assumptions
o Observations are independent = participants do not influence each other
o Dependent variable is normally distributed (pain levels scores)
o Variances are equal in population (can compromise if sample sizes are unequal – levene test
help address this assumption) Dependent samples t test
Objective = whether means of two related groups differ significantly on dependent variable of
interest
o RQ: – Does going to prison have an impact on offending rate after and before?
o Null – Difference on offending rate, before and after going to prison, equal to 00(H : uB - uA = 0)
o Alt – Difference on offending rate, before and after going to prison, NOT equal to 0 0H : uB - uA
≠ 0)
Paired Samples Statistics
Mean N
Pair 1 beforep 56.6963 30
afterp 55.2927 30
Paired Samples Test
Paired Differences
Std. Std. Error Sig. (2-
Mean Deviation Mean t df tailed)
Pair beforep - 1.4036 59.17465 10.80376 .130 29 .898
1 afterp 7
Results
1. Test of significance
T-value = 0.13
P-value = 0.898
If (p value > .05) = Null not rejected = offending rate same before/after going to
prison – sampling error
If (p value ≤ .05) = Null rejected = offending rate differ, before and after going to
prison
o Report mean – greater/lesser? (Before 56, After 55 – after prison offend
less)
o In this sample, p greater, so not different, null not rejected
Effect size = mean difference / std deviation of difference scores
o S (0.2) M (0.5) L (0.8)
o The amount of difference in offending rates between before and after going to prison in terms
of standard deviation units
Eg – 1.4/59.17 = 0.02 – very very small
Assumptions
o Observations are independent within groups= offender do not influence each other
o Scores normally distributed in population One-way ANVOA
Objective = two or MORE groups differ significantly on a on dependent variable of interest
o RQ: does number of words recalled depend on the type of learning strategy used?
o Null – Average Memory scores for A B C is equal in the populatio0 (H : uA = uB = uC)
o Alt – At least one of the learning strategy has a score that is significantly different from the others
Descriptives
wordrecall
N Mean Std. Deviation Test of Homogeneity of Variances
strategy A 10 8.9000 1.52388 wordrecall
strategy B 10 10.2000 1.39841 Levene Statistic df1 df2 Sig.
strategy C 10 4.7000 1.49443 .021 2 27 .980
Total 30 7.9333 2.77841
Multiple Comparisons
ANOVA

More
Less
Related notes for CRIM 320

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.