Marketing Research Ch. 15

Front 1

Prediction

Back 1

Statement of what is believed will happen in the future made on the basis of past experiences or prior observation.

Front 2

Extrapolation

Back 2

One approach to prediction. Detects a past pattern in one variable's behavior and projects it into the future.

Front 3

Predictive model

Back 3

One approach to prediction. Uses relationships among variables to make a prediction.

Front 4

What should predictions be judged upon?

Back 4

Accuracy (goodness).

Front 5

The goodness/accuracy I based on what?

Back 5

Examination of the residuals and errors- comparisons of predictions to actual values.

Front 6

Regression analysis

Back 6

Predictive analysis technique in which one or more variables are used to predict the level of another by use of the straight- line formula.

Front 7

Straight- line formula (2 formulas)

Back 7

y =a + bx

y = B0 + B1X1 + e

Front 8

Bivariate regression analysis

Back 8

Type of regression in which only two variables are used in the regression- predictive model. Has a dependent variable (y) and independent variable (x). (ties the dep and indep variable together)

Front 9

The _______________variable is used to predict the ________________ variable.

Back 9

Independent

Dependent

Front 10

With ___________________________, one variable is used to predict another variable.

Back 10

Bivariate analysis

Front 11

The straight line equation is the basis of __________________________.

Back 11

Regression analysis.

Front 12

Independent variable

Back 12

Used to predict the dependent variable (x in the straight- line equation).

Front 13

Dependent variable

Back 13

That in which is predicted (y in the straight- line equation).

Front 14

Least squares criterion

Back 14

Used in regression analysis; guarantees that the straight line runs through the points on the diagram and is positioned to min vertical distance away from the line of the various points. Best spot.

Front 15

Outliers

Back 15

Data points that are substantially outside the normal range of the data points being analyzed.

Front 16

Standard error of the estimate

Back 16

Used to calculate a range of the prediction made with a regression equation.

Front 17

3 things that must be tested for statistical significance

Back 17

1. Regression model

2. Intercept

3. Slope

Front 18

Regression analysis predictions are estimates that...

Back 18

Have some amount of error in them.

Front 19

The t test is used to determine:

Back 19

Whether the intercepts and slope are significantly different from 0 (the null hypothesis).

Front 20

What does it mean if the computed t values is greater than the table t value?

Back 20

The null hypothesis is not supported.

Front 21

Regression predictions are made with ________________________.

Back 21

Confidence intervals.

Front 22

Multiple regression analysis

Back 22

Uses the same concepts as bivariate regression analysis but uses more than one independent variable.

Front 23

General conception model

Back 23

Identifies independent and dependent variables and shows their basic relationships to one another.

Front 24

4 factors that contribute to purchase intentions to purchase preferences (conceptual model)

Back 24

1. Media exposure, word of mouth

2. Demographics, lifestyle

3. Past behavior, experience, knowledge

4. Attitudes, opinions, feelings

Front 25

What does multiple regression mean?

Back 25

You have more than one independent variable to predict a single dependent variable.

Front 26

Are the x terms in multiple regression additive or multiplicative?

Back 26

Additive.

Front 27

What is the one basic assumption in multiple regression?

Back 27

The inclusion of each independent variable preserves the straight- line assumptions of multiple regression analysis (additivity).

Front 28

Additivity

Back 28

Each new independent variable is added to the regression equation.

Front 29

Regression plane

Back 29

In multiple regression, it is the shape of the dependent variable.

Front 30

What does multiple regression tell us?

Back 30

Which factors predict the dependent variable, which way (the sign) each factor influences the dependent variable, and even how much (the size of b) each factor influences it.

Front 31

2 basic assumption of multiple regression

Back 31

1. A regression plane is used instead of a line.

2. A coefficient of determination indicates how well the independent variables can predict the dependent variable in multiple regression.

Front 32

Independence assumption (multiple regression)

Back 32

The independent variables must be statistically independent and uncorrelated with one another.

Front 33

Variance inflation factor (VIF)

Back 33

Can be used to assess and eliminate multicollinearity.

Front 34

VIF identifies what ________________ should be removed.

Back 34

Independent variables.

Front 35

Any variable with VIF greater than ____ should be removed.

Back 35

10.

Front 36

Another name for multiple regression

Back 36

Coefficient of determination

Front 37

What does multiple regression measure?

Back 37

The strength of the overall linear relationship in multiple regression.

Front 38

Multiple regression ranges from _____ to ______.

Back 38

0 to +1.

Front 39

Multiple regression represents the amount of...

Back 39

The dependent variable that is explained or accounted for by the combined independent variables.

Front 40

Multiple regression findings explain ____% of the dependent variable.

Back 40

.75, 75 %.

Front 41

When is stepwise regression useful?

Back 41

When there are many independent variables and a researcher wants to narrow the set down to a smaller number of statistically significant variables.

Front 42

In stepwise multiple regression, which variables are eliminated?

Back 42

All insignificant independent variables.

Front 43

In stepwise multiple regression, each statistically significant independent variable is added in what order?

Back 43

Order of variance explained.

Front 44

In stepwise multiple regression, what variable is entered into the multiple regression equation?

Back 44

The one independent variable that is statistically significant and explains the most variance.

Front 45

Dummy independent variable

Back 45

Scales with a nominal 0 vs. 1 coding scheme.

Front 46

Standardized beta coefficient

Back 46

Betas that indicate the relative importance of alternative predictor variables.

Front 47

Multiple regression is sometimes used to help a marketer apply _______________________.

Back 47

Market segmentation.

Front 48

3 warnings regarding multiple regression analysis

Back 48

1. It is a statistical tool, not a cause and effect statement.

2. It should not be applied outside the boundaries of data used to develop the regression model.

3. It is complex.

Front 49

Trimmed regression

Back 49

You eliminate the nonsignificant independent variables and then rerun the regression.

Front 50

When should you run trimmed regressions?

Back 50

Iteratively until all betas are significant.

Front 51

What does the resultant regression model express?

Back 51

The salient independent variables.

Front 52

Using standardized betas to compare independent variables allows what?

Back 52

Direct comparison of each independent variable.

Front 53

What is the purpose of using multiple regression as a screening device?

Back 53

To identify variables to exclude.

Front 54

What 4 things are most important when used as a screening device?

Back 54

1. Dependent variable

2. Statistically significant independent variables

3. Signs of beta coefficients

4. Standardized beta coefficients for significant variables

Front 55

5 steps of multiple regression

Back 55

1. Choose indep. and dependent variables.

2. Does linear relationship exist?

3. Is indep. variable statistically significant?

4. Determine the strength of the linear model.

5. Interpret findings.

Front 56

Multicollinearity

Back 56

Violation of the independence assumption that causes regression results to be in error.

Front 57

Independence assumption

Back 57

Statistical requirement that when more than one x variable is used, no pair of x variables has a high correlation.

Front 58

Multiple r

Back 58

(correlation of determination) A number that ranges form 0-1 that indicated the strength of the overall linear relationship in a multiple regression, the higher the better (high indicates it applies well to the scatter points and low means the straight line does not apply well).

Front 59

Beta coefficients and standardized beta coefficients

Back 59

Beta coefficients are the slope values determined by multiple regression for each indep. variable, x. Standardized in the range of .00 to .99 so they can be compared directly to determine their relative importance in y's prediction.

Front 60

Standard error of the estimate

Back 60

Value used to make a prediction at the 95% level of confidence.

Front 61

In what type of analysis is there an underlying general conception model?

Back 61

Regression analysis.

Front 62

Multiple regression equation and what each letter represents

Back 62

y = a + b1x1 + b2x2....+ bmxm

y= dependent variable

x1= independent variable

a= intercept

b1= slope of indep variable

m= # of indep variables in the equation

Front 63

What command is used to run multiple regression?

Back 63

SPACE ANALYZE REGRESSSION LINEAR.

Front 64

It is common to use multiple regression as a ___________________.

Back 64

Screening device.

Front 65

What command is used to do stepwise multiple regression with SPSS?

Back 65

ANALYZE REGRESSION LINEAR.