3.5 Questionnaire Development
3.5.4 Ability Measuring Instruments
3.5.4.1 Subjective/Perceived Fashion Knowledge
The subjective/perceived fashion knowledge instrument measures the respondents’
knowledge about fashion. Respondents are asked to indicate how much they feel they know about fashion (Goldsmith et al., 1996).
The instrument is comprised of nine questions to which the respondents were asked to indicate the extend to which they agreed with every item on five point Likert scales.
After the data was collected, items four, six and eight were reversed due to negative formulations in the questions. Scores were then summed to create composite scores, where higher scores indicated higher subjective/perceived fashion knowledge.
The instrument was adopted to investigate the influence of perceived fashion knowledge on fashion consumption behavior and environmental behavior.
3.5.4.2 Time Resources & Financial Resources
The instruments measure the respondents’ resources in terms of time and money. The first instrument does so by asking respondents how much time they spend on their most recent shopping trip measured in hours and minutes. The second instrument measures the respondents’ financial resources by asking how much the respondent spends on average per month on clothes.
The measurement was included to investigate whether or not financial and time resources have an effect on fashion related behavior and sustainable fashion consumption practices.
3.5.4.3 Skepticism of Environmental Product Claims
The skepticism of environmental product claims instrument was adopted from Hustvedt’s (2006) dissertation on consumer preferences for blended organic cotton apparel.
Hustvedt’s scale was an adoption of a previous measure of skepticism that had been developed and tested by Mohr, Eroglu & Ellen (1998). The instrument measures the respondent’s skepticism of environmental product claims through five questions on environmental product claims and advertisements, which the respondents were asked to rate on a five point Likert scale.
After pretesting the instrument, “environmental” was exchanged with “sustainable” in all statements and questions. Additionally, the terms "social and environmental" were added to give assistance to the comprehension of the respondents. In the Swedish translation a brief definition of what the term “sustainability” entails was added, as this is not a commonly used term in Swedish.
Due to negative question formulation, the first question had to be reversed after the data was collected. The scores were then combined into composite scores, with higher scores indicating higher degrees of skepticism towards environmental product claims.
This scale was relevant to include in this investigation as is looks at skepticism and therefore has a potential to help explain the mechanisms of the relationship between attitudes towards product claims and behavior.
3.5.4.4 Label Knowledge & Label Use
The instrument used in this paper to measure the respondents’ knowledge and use of environmental labels has been adopted from Thøgersen et al.’s (2010) research on consumer responses towards eco-‐labels. The instrument exposes respondents to a series of eco-‐labels and asks to what extent respondents feel they know the given labels.
Next the respondents are asked to which extent they use the same labels when shopping
for fashion items. Both questions are rated on five point Likert scales.
For the purpose of this paper the original scale has been adjusted to fit a clothing context in Sweden and therefore included the following eco-‐labels: the EU Flower, the Bra Miljöval, GOTS, the Nordic Swan and the Oeko-‐tex Standard 100.
The label knowledge and label use instrument was included in the survey in order to explore the effect of labels on behavior and whether or not labels can help promote sustainable behavior.
3.5.4.5 (Objective) Environmental Apparel Knowledge
The (objective) environmental apparel knowledge instrument measures the knowledge that participants possess regarding the impact and consequences of apparel production, use and disposal (Kim & Damhorst, 1998). It does so by asking the respondents to answer 11 questions, which Kim and Damhorst derived from previous studies and literature. In the original study, the authors use a seven point Likert scale. However, for the sake of simplicity the original scale has been reduced to a five point Likert scale for this survey. After pretesting the questionnaire it was also decided that a “Do not know”
category had to be added in the final version.
Items two, three, six and 11 were reversed, as these were false claims. After the data collection had taken place, scores were summed for each respondent and higher composite scores indicated higher objective environmental apparel knowledge.
This measurement was included in this study to investigate the effect of actual knowledge on behavior and thereby see if the focus on knowledge creation and information can be utilized in the promotion of sustainable behavior and practices.
3.5.4.6 Income
Income, even though a demographic variable, was included under the ability construct as it enables the consumer to purchase certain fashion items which would other wise not be available to him or her. It also served as a control variable to check if the sample was in fact representative. In the case of this survey, income was measured by asking the respondent how much monthly personal disposable income they had available. The scale used was a standard scale provided by GfK Sweden.
Income was also measured in order to access if income had any effect on behavior and if income could be a predictor of specific sustainable or unsustainable behavior.
3.5.4.7 Education
As with income, education is a demographic variable. Nevertheless, it also belongs to the ability construct in the MOAB model as it can be of influence on behavior in that education provides the necessary knowledge to understand and thus carry out specific tasks (Thøgersen, 1994).
Education was measured by asking the respondents what their highest completed education was, measuring it on the ISCED scale.
Investigating education was relevant, as previous studies have shown evidence that educational level has en effect on pro-‐environmental behavior (Thøgersen, 2010).
Additionally, education was included as a control variable to check the sample for representativeness.
3.5.5 Behavior Measuring Instruments