W7 Flashcards Preview

Marketing Research Semester 2 > W7 > Flashcards

Flashcards in W7 Deck (34):
1

correlation

tests the degree and direction of 2 variables
r = correlation
1 = positive correlation
0 = zero correlation
-1 = negative correlation

2

pearsons r

calculates the strength of 2 continuous variables, value between -1 and 1

3

p value in pearsons r

the probability that the correlation is really zero in the population

4

linear regression

tests the degree, direction and numeric relationship between two variables

5

regression line equation

y = a + bx
y= dependent variable
a = intercept
b = slope of line
x = independent variable

6

r2

pearsons r squared

7

causality

the relationship between cause and effect

8

linear regression assumptions

sample size of over 50
normality (histograms)
residual error should not increase/ decrease with every change in x (scatterplot)
removal of outliers
variables share linear relationship
homogeneity of residuals

9

multiple regression

tests the degree, direction and numeric relationship between one variable and several independent variables

10

multiple regression equation

y = a + b1x + b2x + b3x

11

variables in correlation

2 continuous

12

variables in simple linear regression

2 continuous

13

variables in multiple linear regression

1 continuous, several independent

14

skewness and kurtosis is only a problem is outside...

-1 and 1

15

cross tabulation and chi-square

tests whether the count in one category is higher then in another

16

variables in chi-square

2 categorical with 2 independent categories

17

assumptions of chi-square

5+ counts in each category

18

what is sig.

p value

19

what is df

degrees of freedom

20

what do you do if the p value is less then the chosen significance level

reject the null hypothesis

21

independent samples t-test

whether the mean scores on one variable differ between two groups at significant levels

22

variables in independent samples t-test

1 categorical (two groups), 1 continuous

23

assumptions of independent samples t-test

normally distributed
homogeneous variables (use Levenes test - large p value)

24

what is the f. statistic

the outcome of Levenes test

25

what is t

the computed test statistic

26

what is the mean difference

the difference between the sample means

27

what happens if the confidence interval contains a 0

the results are not significant

28

paired samples t-test

analyzes the mean score on a single variable differs before and after intervention

29

variables in paired samples t-test

2 continuous (taken at different times)

30

assumptions of paired samples t-test

normally distributed

31

ANOVA

analyzes whether the mean score on a single variable differs between 3+ variables

32

variables for ANOVA

1 continuous, 1 categorical (3+ levels)

33

assumptions for ANOVA

independence
normality
sample size +80

34

multicollinearity

a problem where one variable is linearly predicted from others with a substantial degree of accuracy