exam 3 at a glimpse

Question 1

p-value

Answer 1

significance testing
-tells you whether the group scores are significantly different from each other
- p < 0.05

Question 2

ANOVA statistic

Answer 2

F
- the F value is the ratio of: difference between the groups by difference within group
-F becomes bigger when the variation within groups is small

Question 3

one-way ANOVA df

Answer 3

1. df (between) = Total no. groups - 1
2. df (within) = Total no. of people - Total no. groups
   E.g.
   3 groups
   Total no. of people = 150
   df(between) = 3 -1 = 2 df(within) = 150 - 3 = 147

Question 4

Eta- squared

Answer 4

η2
ANOVA effect size
-How much variance in the DV is explained by the IV
- How big of a difference there is in the groups scores.
- Eta squared
- .01 : small
- .09 medium
- .25: large

Question 5

probability sampling

Answer 5

Probability sampling is useful and gives you the least bias compared to non-probability sampling.

Question 6

sampling bias

Answer 6

there’s something wrong with the way of gathering sample. For instance, if you want to do a survey about middle school student – and then you find out that one day when you go to the middle school, you find out that the low-income students skipped the school more often, so that you find out that the sample that you gathered are mostly students from middle class. Then… you might run into a problem related to sampling bias.

Question 7

sampling error

Answer 7

it’s random. There is no specific reason for this.

Question 8

chi-square statistic

Answer 8

x squared
(like t from t tests and F from ANOVA tests)

Question 9

chi-squared df

Answer 9

• df = (no. of IV categories – 1) * (no. of DV categories -1)
  E.g.
  IV has 2 groups DV has 2 groups Then, df will be?
  (2-1)*(2-1) = 1
  N = sample size (total no. of people in data set)

Question 10

Cramer’s V

Answer 10

Chi-squared effect size
-How big of an effect does the IV have on DV?
-Like r for correlation, Cohen’s d for t tests,
Eta-squared for ANOVA
-In Nominals table > second row
- Range
- .10 small
- .30 medium
- .50 large

Question 11

Factorial ANOVA statistics

Answer 11

-Same p, F and Eta-squared as one-way ANOVA
-Getting a statistically significant result: p< .05 does not mean that all the effects are significant (main 1, main 2, interaction) in an FANOVA (2-way ANOVA). It means that overall the results are significant.

Question 12

Factorial ANOVA df

Answer 12

For Main Effects:
-df(between) = Total no. categories (in the IV) – 1
-For Interaction Effect:
-df (between) = df(IV1) * df(IV2)
-df (within) = Total no. of people (N) - Total no. of groups/cells
E.g. IV1 = 2 levels, IV2 = 3 levels, N = 194 Main Effect 1 df = 2 – 1 = 1
Main Effect 2 df = 3 – 1 = 2 Interaction Effect df = 21= 2 df(within) = 194 – (23) = 188

Question 13

How to determine which test to do:

Answer 13

Determine the combination of IV and DV variable types:
* t test: IV categorical (2 groups) + DV continuous
* One-way ANOVA: IV categorical (3 or more groups) + DV
continuous
* Chi-squared test : IV categorical + DV categorical
* Factorial ANOVA: 2 IVs categorical (2 or more groups) + DV
continuous *categorical (nominal, ordinal) *continuous (interval, ratio)

Question 14

ex ?: Taylor conducted a research to find out whether gender impacted people’s willingness to buy product endorsed by a social media influencer

Answer 14

1-way ANOVA (if more than 2 groups)/ t-test (if 2 groups)

Question 15

ex ?: Claire conducted a research to find out whether race impacted whether people have seen Eternals or not.

Answer 15

Chi-squared test

Question 16

ex ?: As a follow-up study, Claire conducted a research to find out whether there is a significant difference among races and genders in terms of “how many times” each participant saw Eternals

Answer 16

Factorial (2-way)ANOVA