Policy: Why is it critical to consider both the format of your data and the question(s) you wish to answer before selecting a statistical technique to use?
The type of data you collect will determine what tests you can run on it. For instance, you can measure percentages, mode and chi-square for nominal data. You can measure percentile, median, rank-order correlation, or Friedman ANOVA for ordinal data. You can measure mean, standard deviation, product moment correlations, t-tests, ANOVA, regression, or factor analysis for interval data. And you can measure the mean and coefficient of variation for ratio data (Aaker, D., Kumar, V., Day, G., and Leone, R., 2010, p. 295).
The question you are asking will determine what type of statistical technique you use. First state the problem. The steps for hypothesis testing are: 1) clearly state a null and alternate hypothesis, 2) choose the relevant test and a probability distribution, 3) choose the critical value or significance, 4) state the decision rule, 5) collect the data to test the hypothesis, and 6) make a decision and determine if the difference is statistically significant or just normal variation (Aaker, et al., 2010, p. 463). If you wish to determine if a population parameter is higher or lower than a specific value, you would use a one tailed test. If you wish to determine if a population parameter falls between two specified values, you would use a 2-tailed test. If sigma is known, or for very large samples, you would use a z-test. If sigma is not known, or for small samples, you would use a t-test. To determine the effect of an independent variable on a dependent variable, you would use a regression analysis (Aaker, et al., 2010, p. 523). œThe parameter
b
1
indicates that if the variable X is changed by one unit, the variable Y will change by
b
1
units. Thus, if $1 is added to the advertising budget, regardless of the level at which the budget is set, an extra
b
1
customers will be expected to visit the store (Aaker, et al., (2010), p. 527).
Reference:
Aaker, D., Kumar, V., Day, G. and Leone, R. (2010). Marketing research. Hoboken, NJ: John Wiley and Sons
This is a response to the discussion question:
Policy: Why is it critical to consider both the format of your data and the question(s) you wish to answer before selecting a statistical technique to use?
I need a page here (double spaced)
Please make sure to write the response as if I was writing it. I’ve given you examples
Data format and question implications
Note # 1
by Kathleen Lee Myers
March 27, 2014 8:08 PM
_______________________________________________________________________________
This is a response to the discussion question:
Policy: Why is it critical to consider both the format of your data and the question(s) you wish to answer before selecting a statistical technique to use?
I need a page here (double spaced)
Please make sure to write the response as if I was writing it. I’ve given you
Conference 7
Note # 2
by Susan Smith
______________________________________________________________
Aaker, D.A., Kumar, V., Day, G.A. & Leone, R.P. (2010). Fundamentals of data analysis (pp. 439-459). Marketing Research (Tenth Edition). Hoboken, NJ: John Wiley & Sons, Inc.
This is a response to the discussion question:
Policy: Why is it critical to consider both the format of your data and the question(s) you wish to answer before selecting a statistical technique to use?
I need a page here (double spaced)
Please make sure to write the response as if I was writing it. I’ve given you examples
March 28, 2014 11:00 AM