Week 12 Flashcards
What’s the before and after with control group design
True experimental design that involves random assignments of subjects or test units to experimental and control groups and pre and post measurements of both groups
Why is the difference between the pre and post measurements of the control group important
Provide a good estimate of the effect of all extraneous influences experienced by each group
What is after only with control group designs
The true experimental design that involves random assignment of subjects or test units to experimental and control groups but no pre measurement of the dependent variable
What is quasi experiments
Where the researcher lacks complete control over the scheduling of treatments or must assign respondents to treatments in a non random manner
What is interrupted time series design
Repeated measurements of an effect “interrupts” previous data patterns
Why does time series experimental designs have greater interpretability
The may measurements allow more understanding of the effects of extraneous variables
What are the two fundamental weakness of the time series design
- Inability to control history
- Possibility of interactive effects of testing and evaluation apprehension resulting from repeated measurements taken on test units
What is multiple time series design
Interrupted time series design with control
What are test markets
Real world testing of a new product or some element of the marketing mix use experimental or quasi experimental design
What are test market studies designed to provide information on which issues?
- Estimates of market share and volume
- Effects that the new product will have on sales of similar products already marked by the company called cannibalization rate
- Characteristics of consumers who buy the product
- Behaviour of competitors during the test
What are the four types of test markets
- Tradtional
- Scanner or electronic
- Controlled
- Stimulated
What traditional or standard markets
Involves testing the product or other elements of the marketing mix through regular channels of distribution
What are scanner or electronic test markets
Markets where research firms have panels of consumers who carry cards for use in buying products
What are controlled test markets
Managed by research suppliers who ensure that the product is distributed through the agreed upon types and numbers of outlets
Why are test markets be the last step of a research process
Test markets are expensive
Why are test markets important
- Provides a vehicle where the firm can obtain a good estimate of a products sales potential under realistic market conditions
- Test should identify weaknesses of the product and the proposed marketing strategy for the product and give management an opportunity to correct those weaknesses
What are hypothesis test of proportions
Test to determine whether the difference between proportions is greater than would be expected because of sampling error
How does hypothesis test of proportions work
- Specify the null and alternative hypothesis
- Specify the level of sampling error
- Calculate the estimated standard error, using the P value specified in the null hypothesis
- Calculate the test statistics
Calculate the test statistics
- Specify the null and alternative hypothesis
- Specify the level of sampling error allowed
- Calculate the estimated sd using the value of p specified in a null hypothesis
- Test statistic
When do we reject null hypothesis
Z-value is larger than the critical z-value
What are the specifications required and the procedure for testing hypothesis
- Null hypothesis
- Alternative hypothesis
How do you test two proportions in independent samples
- Specify the null and alternative hypothesis
- Set the level of sampling error alpha
- Calculate the estimated standard error of the differences between the two proportions
- Calculate the test statistic
- State the result
What are analysis of variance
Testing the differences among the means of two or more independent samples
What are one way anova
Used to analyze experimental results
How do you anova
- Specify the null and alternative hypothesis
- Sum the squared differences between sample mean and the overall sample mean weighted by sample size
- Calculate the variation among groups measured by the mean sum of squares among groups
- Sum the squared differences between each observation
- Calculate the variation within the sample groups as measured by the mean sum of squares within groups
- Calculate the F statistic
- State the results
What is the f-test
Tests of the probability that a particular calculated value could have been due to chance
What is the p-value
- Exact probability of getting a computed test statistic that is due to chance
- Smaller the p-value the smaller the probability that the observed results occurred by chance
What are bivariate techniques
Statistical methods of analyzing the relationship between two variables
What are independent variables
Variables believed to affect the value of the dependent variable
What are dependent variable
Variables expected to be explained or caused by the independent variable
What are bivariate regression analysis
A statistical procedure appropriate for analyzing the relationship between two variables when one is considered the dependent variable and the other independent variable
What is a scatter diagram
Graphic plots of the data with dependent variables on the Y axis and independent variables on the x and shows the nature of the relationship between two variables, linear or non linear
What are the least squares estimation procedures
The least squares procedures are fairly simple math techniques that can be used to fit data for x and y to a line that be represents the relationship between two variables
What is the regression line
Predicted values for Y based on a and b
What is the strength of association
Estimated regression function that describes the nature of the relationship between x and y
What is coefficient of determination
Percentage of the total variation in the dependent variable explained by the independent variable
What is statistical significance of regression results
This is the total variation in y that is two components sum of squares of explained variation and un explained variation
What is sum of squares due to regression
Variation explained by the regression
What are error sum of squares
Variation not explained by regression
