VICTORIA UNIVERSITY
School of Economics and Finance
BEO2254 STATISTICS FOR BUSINESS AND MARKETING |
Guidelines for Presenting Lab Exercises
Intent
This unit attempts to do more than simply teach students how to perform a range of statistical tasks using SPSS. It also seeks to develop communication skills by requiring interpretation and discussion of the results in documents along the lines of short business reports.
Formatting Reports
Up to a point, what constitutes good style in writing a report is a matter of taste, however this is not to say that anything at all is satisfactory. The requirements of correct spelling and grammar are, of course, not negotiable. Students who ignore the spelling and grammar checker in Word can expect to be penalized.
The use of Microsoft Office is an effective means of formatting graphs and tables, but students may wish to format SPSS output within the SPSS program by changing the default settings. Using Office, however, is likely to produce better results and with less effort.
Unless the default settings in SPSS are changed, in general, inserts such as tables and graphs will have inappropriate formats for report documents. In particular, the font is unlikely to match the rest of the document, tables will have too many grid lines and in some instances tables will contain too much information. Further, it may be difficult to make the edges of an insert align with the margins of the document.
1. Use a front page. (Show subject name and code, student name and code, the name of the document and not much else.)
2. Set all margins at a minimum of 2.5 cms.
3. For all but the front page, write in Times New Roman 12 (or something very similar), use space and a half, and use left and right justification. For the front page, use 36 and/or 48 point Times New Roman with centre or right hand justification.
4. For tables use Word Table AutoFormat Simple 1 to present results. Edit out any unnecessary material. Consider using bold and/or bold italic for headings within tables. If you elect to use graphs, ensure that they conform to the rest of the document. Make all inserts conform to the margins of the document. Use a table numbering system.
5. Consider using an “Executive Summary” approach. I.e. break up the report with headings and sub-headings and make extensive use of bullet points. (Everyone should know how to use the Headings and Captions in Word, but if you don’t, now is the time to learn.)
6. On no account should you be required to reproduce the SPSS results using formulas, statistical tables and manual calculations. The only time you may need to perform a manual calculation for an exercise is when there is a task that SPSS cannot perform. Nevertheless, when calculations are required (not often) or when you think it is appropriate to insert a formula, do not write it in with a pen. Invest a little time (about 10 minutes) learning how to use Insert/Object/Microsoft Equation 3.0 (and switch off Float over text).
7. Place the document in a plastic spiral cover.
8. Attach a Standard Assignment Cover Sheet available on www.businessandlaw.vu.edu Place it at the end of the document.
9. Most importantly, the exercise documents should be stylish.
Formatting SPSS Inserts
In what follows you will see the output from three SPSS procedures as they appear in the SPSS output viewer. Generally these tables are unsuitable for inclusion in documents and accompanying them are edited versions that are more suitable. These edited versions, however, are not only acceptable formats.
Cross-tabulations
The following was obtained using Copy objects from SPSS to Word.
If the box appears empty, switch to Print Preview or Print.
The following was obtained using Copy from SPSS to Word and editing after applying AutoFormat Simple 1.
Table 1: Cross-tabulation of Type of Store by City
|
|
Type Of Store |
Total |
|
|
Supermarket |
Other |
|
City |
Melbourne |
10 |
10 |
20 |
|
Adelaide |
7 |
13 |
20 |
|
Brisbane |
8 |
11 |
19 |
|
Sydney |
11 |
10 |
21 |
|
Perth |
10 |
10 |
20 |
|
Total |
46 |
54 |
100 |
Some might not like the italic, however the edited table is consistent with the style of modern journals and reports.
t - Tests
The following was obtained using Copy objects from SPSS to Word (after deleting a few cells in SPSS).
The following was obtained using Copy from SPSS to Word and editing after applying AutoFormat Simple 1.
Table 2: Test of Equality of Mean Sales at Supermarkets and Other Stores
Type of Store |
n |
Mean |
Levene’s Test for Equality of Variances |
Test for Equality of Means |
Supermarket |
16 |
107.50 |
F = 0.42[.524] |
t28 = 1.86[.074] |
Other |
14 |
98.86 |
|
|
Note that a considerable amount of information produced by SPSS has been deleted. The degrees of freedom are shown as a subscript and the significance or p values have been shown in square brackets immediately after the statistics from which they are derived. The table should be accompanied by sentences such as the following.
Based on a random sample of 30 stores we are unable to reject the null hypothesis at the 0.05 level of significance that the mean sales at supermarkets and others stores are equal. The results of the test are shown in Table 2.
Regressions
The following was obtained using Copy objects from SPSS to Word.
The following was obtained using Copy from SPSS to Word and editing after applying AutoFormat Simple 1.
Table 3: Cross-sectional Demand Model
Dependent Variable: SALES |
|
Coefficient |
t[Sig.] |
(Constant) |
2056.0 |
0.91[0.367] |
PRICE |
-692.0 |
-3.00[0.005] |
INCOME |
199.7 |
17.81[0.000] |
POPULATION |
235.2 |
6.96[0.000] |
|
|
|
|
|
|
|
|
|
|
But consider the formula for the standard error using a stratified random sample. It simply cannot be produced and look good without using the equation editor.
If the equation box is blank, click print preview or print a hard copy. If you want to see how the formula has been constructed from the equation templates double click anywhere in the box.
Remember that you will not have to write many equations, but if you do, they should look professional