Experiment Design

Initially, the research focused on those respondents who have online shopping experience.

Respondents learned that they would participate in one study about mood states of shopping decisions. To induce respondents to form initial mood states before purchasing, respondents were randomly assigned to two mood-inducing conditions, to complete the mood scales. Then, an online shopping scenario was designed to stimulate them into impulse buying behaviors, and the impulse buying scale was used to check whether they were spurred on to buy impulsively. Last the respondents were randomly assigned into three scenarios about numbers of reason provided,

namely none (N=0), less (N=1), more (N=5), and their satisfaction was analyzed. In addition, a regret scale, which was cited by Schwartz (2004), was used to examine the influence of tendency to regret on the relationship between reason provided and satisfaction. Post-purchase satisfaction was analyzed in particular for the effect of the variable. After the experiment, participants were thanked for their help.

3.2.2 Stimulus Development

Manipulation of mood prior to purchase

To manipulate moods, the study selected some pictures for each mood state. There were four interesting and humorous comedies for positive mood manipulation, and four pictures about butchering seals selected for negative mood manipulation (APPENDIX 2). For the purpose of a manipulation check, participants were asked to report their mood using the mood scale adapted from Swinyard (1993) (sad/happy, bad mood/good mood, irritable/pleased, depressed/cheerful).

Manipulation of Impulse Buying

To stimulate respondents into impulse buying behaviors, an online shopping scenario was designed to achieve the purpose. First, the study referred to the product categories from the most popular shopping website (InsightXplorer, 2005), Yahoo! Shopping Mall, to form a virtual shopping environment (APPENDIX 1), and let respondents plan their own shopping lists according to their disposable incomes. Second, in light of Beatty and Ferrell’s (1998) definition of impulse buying (not including reminder items), an appealing purchase alternative was selected, KINGMAX flash drive. This was on sale, to attract respondents to buy the product after they had made their shopping lists (APPENDIX 1). Once respondents chose the impulse-inducing item as

their purchase decisions, they were identified as impulsive buyers. The sample without constructing impulse buying behaviors was eliminated. Finally, the study used the scale of impulse buying of Beatty and Ferrell (1998) to check whether the manipulation was successful (e.g. “When I bought flash drive, I felt a spontaneous urge to buy it.”, scale from 1 = strongly disagree to 7 = strongly agree).

Manipulation of Reason Provided

Reason provided would have the effect of comfort and persuasiveness to influence satisfaction, mentioned in a previous discussion. Here the author use some buying reasons and assigned them into three conditions, namely: no reason (N=1), less reasons (N=1), and more reasons (N=5), N means the numbers of reasons provided from neutral sources. The contents of reason provided are shown in Table 3-2 (APPENDIX 3):

Table 3- 1 Reason Provided Table Numbers (N) of

Reason Provided

Description

None (N=0) No reason was provided.

Less (N=1) 9 The flash drive is useful and/or can be applied in the future.

More (N=5)

9 The flash drive is useful and/or can be applied in the future.

9 A famous flash drive is on sale! Of course you must buy it immediately!

9 The flash drive looks pretty charming! It is handy and special. This will be a good buying decision!

9 It is too much of a waste of time and energy to return goods, so…accept it!

9 You bought the flash drive just for fun, now you are happy anyway, so it is worthwhile!

3.3 Measures 3.3.1 Mood State

A mood scale adapted from Swinyard (1993) was used to check the effect of the mood manipulation. It measures mood at a particular point of moment on a simple good/bad continuum rather than attempting to assess various dimensions of mood (Bruner, Hensel, & James, 2005).

The scale consisted of four summed bipolar measures coded in a seven-point scale and the alpha was originally.85. Our Cronbach’s α of the experiment was 0.974.

Scale Items:

1. sad / happy

2. bad mood / good mood 3. irritable / pleased 4. depressed / cheerful

3.3.2 Impulse Buying

The scale called impulsivity by Beatty and Ferrell (1998) is composed of five, seven-point Likert-type statements measuring the extent to which a consumer indicates experiencing a strong urge to spontaneously make a particular purchase without hesitation or consideration of the consequences (Beatty & Ferrell, 1998). The scale developed in dissertation research by Jeon (1990) was unidimensional and they showed validity with a coefficient alpha of .70. A mean score of 4 or above represented impulsivity. Our Cronbach’s α of the experiment was 0.758.

Scale Items:

1. When I bought the flash drive, I felt a spontaneous urge to buy it.

2. When I bought the flash drive, I felt I would not be able to get it off my mind until I bought it.

3. When I bought the flash drive, I really did not consider the consequences of the purchase.

4. When I bought the flash drive, I did it without much hesitation.

5. When I saw the flash drive, I could not resist it.

3.3.3 Satisfaction

The scale is composed of twelve Likert-type items and measures consumers’ degree of satisfaction with a product they have recently purchased (Bruner et al., 2005). It was originally generated and used by Westbrook and Oliver (1981) measure consumer satisfaction with cars but Mano and Oliver (1993) appear to have adapted it so as to be general enough to apply to

whatever a respondent was considered. The research followed Oliver and Swan (1989) to use a seven-point format with their alphas of .94. Our Cronbach’s α of the experiment was 0.870 after the validity was qualified as required.

Scale Items:

1. This is one of the best flash drive I could have bought.

2. This flash drive is exactly what I need.

3. This flash drive hasn’t worked out as well as I thought it would. ® 4. I am satisfied with my decision to buy this flash drive

5. Sometimes I have mixed feelings about keeping it. ® 6. My choice to buy this flash drive was a wise one.

7. If I could do it over again, I would buy a different make/model. ® 8. I have truly enjoyed this flash drive.

9. I feel bad about my decision to buy this flash drive. ® 10. I am not happy that I bought this flash drive. ® 11. Owning this flash drive has been a good experience.

12. I am sure it was the right thing to buy this flash drive.

3.3.4 Tendency to Regret

The article used a regret scale which is from the American Psychological Association to evaluate the levels of regret. After summing up all the scores of the questions, the scale shows that the higher the total amount, the easier people feel regret (Schwartz, 2004). The point is from 1 (strongly disagree) to 7 (strongly agree). Our Cronbach’s α was .711 after the validity was qualified as required.

Scale Items:

1. Once I make up my mind, I won't look back again

2. Every times I make a decision, I always want to know what would have happened if I had made a different choice then.

3. Even though I have made a good decision, I still feel I have lost if a better choice is found.

4. I collect related information of all alternatives before making a decision.

5. When I recall my performance, I usually take a missed opportunity to heart.