VIEWS: 188 PAGES: 51 CATEGORY: Personal Finance POSTED ON: 7/10/2010
Executive Summary Samples document sample
SPSS Tutorial (2) Research Methods & Data Analysis Assignment example Methodology (about 200 words) Results (about 400 words) Discussion (about 400 words) Executive summary (max 200 words – structured into bullet points) APPENDIX – including a small selection of tables, graphs and statistical output NOTE: the following examples of the final report are purely imaginary!!! They are not based on actual results!!! Methodology section • State the objective • Briefly describe the methodology you chose: – What does the methodology do? – Why is it useful for your objective Methodology (example) The objective of this study was to test whether being vegetarian influences the amount spent in GCS stores. An independent sample t-test was chosen as the methodology to investigate such hypothesis. The t-test allows to compare two means from different consumer groups and test the null hypothesis that the two means are equal. If the probability of the t-statistics falls below a threshold level (set at 0.05, i.e. 5%) then the null hypothesis is rejected in favour of the alternative. The t-test is based upon the normal distribution of the target variable (I.e. the amount spent) within each of the groups. If the sample size is reasonably large (>40-50 units) it is possible to exploit the normal approximation. […] 118 words Results • Show/summarise the most relevant output of your SPSS analysis • Describe and comment statistically (i.e. not in marketing terms) such output • Illustrate any limit/problem that might have emerged from your application Results (example) The t-test was carried out to check whether the following customer characteristics led to statistically significant differences in the group means: – Vegetarian – Use coupon – Gender Table 1 summarises the output of the Independent Samples T-test for the 3 above customer characteristics [Table 1 here] Table 1 shows that the t-statistic for the vegetarian characteristic has a p-value of 0.77. As this p-value is above 0.05, the null hypothesis of equal means can not be rejected. This means the vegetarian factor is not influencing the amount spent. However, the mean comparison hypothesis test does not take explicitly into account the potential influence of other disturbing factors (e.g. store size). Partial correlation and regression analysis could give further information in that direction, but for the objective of this study and given the very high p-value we can confidently assume that the t-test result are reliable. […] Discussion • Now interpret the result you have presented in the previous section under an operational (marketing) perspective, leave out technicalities and focus on the main findings. Discussion (example) This study showed that being vegetarian is not an influential factor in determining the amount spent, while there are significant differences in terms of gender and the use of coupon. More specifically, table 2 gives the average amount spent for male/female and user/non users of coupon. It looks that the amount spent by men is significantly higher, and also the use of coupon lead to a higher expenditure. This could lead to a strategy for increasing the amount spent as follows: • Create an advertising campaign to attract males into the GCS chain • Improve the distribution of coupon (after an accurate cost/benefit analysis) Executive summary • Now just extract a bullet point list summarising the key passages of your study. Executive summary (example) • The objective of the study is to investigate what factors influence the amount spent • We used hypothesis testing (independent samples t-test) as a methodology • Other methodologies (ANOVA, partial correlations) could give further indication • Being vegetarian is not a relevant factor • Gender and use of coupon are relevant factor • An male-targeted advertising strategies and the calibration of the distribution of coupon could increase the profits for GCS Task A • Examine the relationship between the amount spent and the following customer characteristics: – Being male/female – Being vegetarian – Shopping for himself / for himself and others – Shopping style (weekly, bi-weekly, etc.) Potential methods: • Battery of hypothesis testing & Analysis of variance • Correlation / Regression Analysis • B Task between the amount Examine the relationship spent and the following customer characteristics: – Hypothesis: the average amount spent in health- oriented shop is higher than those of other shops. True or false? – Test the same hypothesis accounting for different shop sizes Potential methods: • Battery of hypothesis testing & Analysis of variance • Partial correlation (accounting for size) • Regression Analysis • Task C average amount Find a relationship between the spent per store and the following store characteristics: – Size of store – Health-oriented store – Store organisation Potential methods: • Transform the customer data set into a store data set • Battery of ANOVA • Correlation / Regression Analysis • Task D those that use Hypothesis: is the amount spent by coupon significantly higher? • What is the most effective way of distributing coupons: – By mail – On newspapers – Both Potential methods: • Recode the variable into 1=not using coupon and 2=using coupon • Hypothesis testing • Analysis of variance SPSS basics • Opening SPSS files • Defining variables • Restructuring data • Saving data • The output window • Cross-tabulation • Graphs Variable view Data view Case summaries • Analyze / Report / Case summaries – Select target variable(s) – Select grouping variable(s) – Include additional statistics Variable(s) you are interested in Grouping variables Do not limit/display cases Click here to choose the statistics you need Output window Categorising variables • Transform/categorize variables – Select variable – Choose number of categories Computing new variables • Transform/Compute – Choose expression – Define “if” category Define a Name the new numeric variable expression to compute it Define a condition If you want to work with ‘stores’ as rows • Data / Aggregate Aggregating variable Name the new file Descriptive statistics in SPSS • Click on Analyze / Descriptive Statistics / Frequencies • Select the variable you are interested in • Select the STATISTICS you are interested in First select the variable Then choose the statistics SPSS output Statistics Amount spent N Valid 779 Missing 0 Mean 404.4871 Std. Error of Mean 4.13528 Median 394.0800 a Mode 274.70b Std. Deviation 115.41804 Percentiles 10 262.4020 c 20 310.5520 30 335.6920 40 366.2010 50 394.0800 60 427.1540 70 455.4820 80 497.3320 90 556.5080 a. Calculated from grouped data. b. Multiple modes exist. The smallest value is shown c. Percentiles are calculated from grouped data. Charts & descriptive stats Charts Amount spent 80 60 40 20 Frequency Std. Dev = 115.42 Mean = 404.5 0 N = 779.00 10 15 20 25 30 35 40 45 50 55 60 65 70 75 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Amount spent Grouped statistics in SPSS • Click on Analyze / Custom Tables / General tables • Select the target variable(s) as rows • Select the grouping variable(s) as column • Choose the statistics you want to be computed Tick here to have the statistics in the table Grouping variable Select the statistics you want Output Amount spent Standard Mean Median Mode Percentile 10 Percentile 95 Error of Mean Std Deviation Valid N Size of Small 391.55 380.75 274.70 241.48 635.48 10.36 121.21 N=137 store Medium 399.53 388.12 106.03 259.92 625.23 5.99 117.15 N=383 Large 418.67 416.59 170.03 288.17 626.61 6.74 108.53 N=259 Hypothesis testing in SPSS • One-sample test (value of the mean in the population) • Analyze / Compare Means / One sample test Click on OPTIONS to choose the confidence level Output T-Test One-Sample Statistics Std. Error N Mean Std. Deviation Mean Amount spent 779 404.4871 115.41804 4.13528 One-Sample Test Test Value = 400 95% Confidence p value Interval of the Mean Difference t df Sig. (2-tailed) Difference Lower Upper Amount spent 1.085 778 .278 4.4871 -3.6305 12.6047 The null hypothesis is not rejected (as the p-value is larger than 0.05) Test on two means (independent samples) • Analyze / Compare means / Independent samples t-test Specify which groups you are comparing Output Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Mean Std. Error Difference F Sig. t df Sig. (2-tailed) Difference Difference Lower Upper Amount spent Equal variances 1.244 .265 -2.270 394 .024 -27.1146 11.94454 -50.59764 -3.63162 assumed Equal variances -2.194 251.912 .029 -27.1146 12.35795 -51.45270 -2.77657 not assumed The null hypothesis is rejected (as the p-value is smaller than 0.05) • Analyze / ANOVA in SPSS Compare means / One-way ANOVA ANOVA dialog box ANOVA output ANOVA Total output Sum of Squares df Mean Square F Sig. Between Groups 309.600 2 154.800 14.221 .000 Within Groups 293.900 27 10.885 Total 603.500 29 Correlation and covariance in SPSS Choose between bivariate & partial Bivariate correlation Select the variables you want to analyse Require the significance level (two tailed) Ask for additional statistics (if necessary) Bivariate correlation output Correlations Shopping style Use coupons Amount spent Shopping style Pearson Correlation 1 .157** .159** Sig. (2-tailed) . .000 .000 N 779 779 779 Use coupons Pearson Correlation .157** 1 .291** Sig. (2-tailed) .000 . .000 N 779 779 779 Amount spent Pearson Correlation .159** .291** 1 Sig. (2-tailed) .000 .000 . N 779 779 779 **. Correlation is significant at the 0.01 level (2-tailed). Partial correlationsList of variables to be analysed Control variables Partial correlation output - - - P A R T I A L C O R R E L A T I O N C O E F F I C I E N T S - - - Controlling for.. SIZE STYLE Partial correlations still AMTSPENT USECOUP ORG measure the correlations AMTSPENT 1.0000 .2677 -.0116 between two variables, but ( 0) ( 775) ( 775) eliminate the effect of other P= . P= .000 P= .746 variables, i.e. the correlations are computed on consumers USECOUP .2677 1.0000 .0500 shopping in stores of identical ( 775) ( 0) ( 775) size and with the same P= .000 P= . P= .164 shopping style ORG -.0116 .0500 1.0000 ( 775) ( 775) ( 0) P= .746 P= .164 P= . (Coefficient / (D.F.) / 2-tailed Significance) " . " is printed if a coefficient cannot be computed Bivariate regression in SPSS Regression dialog box Dependent variable Explanatory Leave this variable unchanged! Regression output Coefficientsa Unstandardized Standardized Coefficients Coefficients Model B Std. Error Beta t Sig. 1 (Constant) 140.359 34.715 4.043 .001 Age 4.577 .838 .807 5.464 .000 a. Dependent Variable: Cholesterol (mg/100 ml) Value of the Statistical coefficients significance Is the coefficient different from 0? Multivariate regression in SPSS • Analyze / Regression / Linear Simply select more than one explanatory variable Output Coefficientsa Unstandardized Standardized Coefficients Coefficients Model B Std. Error Beta t Sig. 1 (Constant) 296.482 19.792 14.980 .000 Health food store 9.721 15.012 .024 .648 .517 Size of store 9.753 6.070 .059 1.607 .109 Gender -69.598 7.483 -.302 -9.301 .000 Vegetarian -1.910 12.570 -.005 -.152 .879 Shopping style 22.760 6.069 .123 3.750 .000 Use coupons 30.417 3.512 .285 8.662 .000 a. Dependent Variable: Amount spent How good is the model? Model Summary Adjusted Std. Error of Model R R Square R Square the Estimate 1 .439 a .193 .187 104.08167 a. Predictors: (Constant), Use coupons, Vegetarian, Gender, Health food store, Shopping style, Size of store • The regression model explain less than 19% of the total variation in the amount spent