Question

Studies have shown that the frequency with which shoppers browse Internet retailers is related to the frequency with which they actually purchase products and/or services online. The following data show respondents age and answer to the question “How many minutes do you browse online retailers per week?”

Age Time

13 5662

19 4549

16 3772

44 1872  

32 2799

52 1355

39 1966

15   5682

40 1602

53 1186

48 1832

37 2253

36 2241

42 1001

30 2474

42 1943

28 3021

11 5682

32 2192

39 1784

23 2707

37 1801

17 4827

11 2693

18 4340

50 1399

52 1593

9 9154

41 1504

26 2627

30 2575

32 2711

53 2368

1. Use Data > Data Analysis > Correlation to compute the correlation checking the Labels checkbox.

2. Use the Excel function =CORREL to compute the correlation. If answers for #1 and 2 do not agree, there is an error.

3. The strength of the correlation motivates further examination. a) Insert Scatter (X, Y) plot linked to the data on this sheet with Age on the horizontal (X) axis. b) Add to your chart: the chart name, vertical axis label, and horizontal axis label. c) Complete the chart by adding Trendline and checking boxes

4. Read directly from the chart a) Intercept = b) Slope = c) R2 = Perform Data > Data Analysis > Regression.

5. Highlight the Y-intercept with yellow. Highlight the X variable in blue. Highlight the total standard error in orange SUMMARY OUTPUT

6.Use Excel to predict the number of minutes spent by a 37-year old shopper. Enter = followed by the regression formula. Enter the intercept and slope into the formula by clicking on the cells in the regression output with the results.

7. Is it appropriate to use this data to predict the amount of time that a 68-year-old will be on the Internet? If yes, what is the amount of time, if no, why?

Answer 1

1.

2.

3.

4. a) intercept = 6223.04

b) slope = -103.27

c) R² = 0.65

5.

6.

Hence the predicted number of minutes = 2402

7. No we cannot predict since the value is out of scope.