27.10.2012 Views

Analysis of Sales Promotion Effects on Household Purchase Behavior

Analysis of Sales Promotion Effects on Household Purchase Behavior

Analysis of Sales Promotion Effects on Household Purchase Behavior

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Table 6.1 c<strong>on</strong>tains the results <str<strong>on</strong>g>of</str<strong>on</strong>g> several 2-sample independent t-tests, each testing<br />

the null hypothesis that two populati<strong>on</strong> means are equal, based <strong>on</strong> the results observed in<br />

two independent samples. In this situati<strong>on</strong>, testing whether the two group (the 200<br />

households that are a member <str<strong>on</strong>g>of</str<strong>on</strong>g> the final analysis set versus those 3860 households who<br />

are not) means are equal for several variables.<br />

Table 6.1: Results independent t-test<br />

110<br />

Variable Observed-two-tail<br />

significance level<br />

Representative<br />

Social class 0.08 YES<br />

Size <str<strong>on</strong>g>of</str<strong>on</strong>g> the household 0.07 YES<br />

Age <str<strong>on</strong>g>of</str<strong>on</strong>g> the housewife 0.96 YES<br />

<strong>Household</strong> Cycle 0.62 YES<br />

Yearly total number <str<strong>on</strong>g>of</str<strong>on</strong>g> grocery shopping<br />

trips<br />

0.00 NO<br />

Yearly total amount <str<strong>on</strong>g>of</str<strong>on</strong>g> grocery expenditures 0.43 YES<br />

The final analysis sample <str<strong>on</strong>g>of</str<strong>on</strong>g> households used in this dissertati<strong>on</strong> seems to differ from the<br />

entire household panel with respect to the total number <str<strong>on</strong>g>of</str<strong>on</strong>g> shopping trips. The households<br />

in the analysis set make (significantly) less shopping trips (<strong>on</strong> average 39 compared to 45<br />

shopping trips per year for the remaining 3860 households). Therefore, frequent shoppers<br />

are underrepresented in the analysis set <str<strong>on</strong>g>of</str<strong>on</strong>g> households. More frequent shoppers could have<br />

the tendency to visit more different stores, resulting in a smaller probability for having a<br />

primary store.<br />

A more in-depth study revealed that the difference is caused by a small percentage<br />

<str<strong>on</strong>g>of</str<strong>on</strong>g> very frequent shoppers (2 or even 3 shopping trips per day <strong>on</strong> average) in the entire<br />

household panel, which is underrepresented in the analysis sample. When excluding the top<br />

10% <str<strong>on</strong>g>of</str<strong>on</strong>g> most frequently shopping households from both the entire panel and the analysis<br />

sample, the (two-sample independent) t-statistic does c<strong>on</strong>clude that the average numbers <str<strong>on</strong>g>of</str<strong>on</strong>g><br />

shopping trips are comparable. So the bulk (90%) <str<strong>on</strong>g>of</str<strong>on</strong>g> shoppers in the analysis sample is

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!