Data collection

A total of six independent data samples were collected in four experimental settings to provide an understanding of the phenomenon of prosocial status signaling via organic food consumption. Table 1 reveals the details of the collected data samples. It is noteworthy that Datasets 1 and 2, as well as 4 and 5, were collected with practically identical methods (i.e., in the analyses, they can be treated as one broad sample). All data were collected as a part of a multidisciplinary research project funded by the Academy of Finland:

Mainstreaming Green Economy – Legitimation of organic production and consumption (MainGreen). The partners in this consortium came from the Universities of Vaasa, Helsinki (Ruralia Institute), and Jyväskylä (JSBE), including consumer researchers, scholars working with rural developing issues, and critical management scientists.

Table 1. Key features of the collected samples

*= all study participants had graduated from or were enrolled in a university

Data 1 Data 2 Data 3 Data 4 Data 5 Data 6

3.3.1 Procedure for Experiment 1: Article 1

Two separate datasets were collected by altering the social visibility of the choice situation, in the university library of a large city, on the pretext of an academic memory recall task. The time difference between the data collections was approximately two months. The first collection (data 1, private choice) consisted of 80 student consumers (mean age 26.1 years, 50% men), while the second (data 2, public choice) consisted of 88 student consumers (mean age 28.3 years, 50%

men). In both settings, the students were randomly assigned to one of two conditions: 1) status-priming under private condition and 2) control-priming under private condition or 1) status-priming under public condition and 2) control-priming under public condition. Because the variables were otherwise identical, in the subsequent analysis, the datasets were pooled.

Status motives (vs. control motives) were triggered by showing participants a word list at the beginning of the study (20 Finnish nouns) that contained either 12/20 words related to high status (e.g., luxury product, designer watch, and first class) or a list containing only neutral control words (e.g., backpack, table lamp, and fraction). The participants had three minutes to memorize the words: they were told that they would be asked about the words again at the end of the study (see Maio et al., 2009). Data collectors ensured that the words were memorized in the time required. After that but before they were allowed to make their product choices, the participants had to answer certain filler questions related to their use of technology. This section was added to conceal the true purpose of the study.

Then, the study participants had to make dichotomous choices between six counterbalanced product pairs. Two pairs included an organic option: bacon and coffee (see Chapter 3.2). The social visibility of the choice situation was manipulated by instructing some of the participants, before the product choice task (public setting, Dataset 2), that they should imagine that they are shopping for ingredients with a friend for an important dinner (cf. Griskevicius et al., 2010;

Ordabayeva & Chandon, 2010), while other participants were simply asked to choose between the alternatives (private setting, Dataset 1). After that, their activity levels (e.g., mood and tiredness), production method, and brand attitudes were investigated, along with their background characteristics.

Lastly, participants were asked to indicate, both orally and in writing, what they believed the actual purpose of the study to have been. These post-study interviews did not reveal any suspiciousness. The procedure took an average of 15 to 20 minutes. The capability of the status word list (relative to the control word list) to elicit a desire for status among consumers was verified with a separate manipulation check, specifically applying Eastman et al.’s (1999) status consumption statements. Similarly, the capability of the selected organic food products to confer prosociality was successfully pre-tested with a separate sample.

More detailed descriptions of the procedures and conducted pre-tests can be found in Article 1.

3.3.2 Procedure for Experiment 2: Article 1

A total of 257 student consumers (mean age 25.0 years, 45% men) participated in the study in a university campus area of a large Finnish city; the study was presented as a memory recall task, which also involved tasting food samples (Dataset 3). Participants were randomly assigned to one of four conditions: 1) status-priming under private condition, 2) control-priming under private condition, 3) status-priming under public condition, and 4) control-priming under public condition.

The design of the experiment was otherwise identical to that of Experiment 1, but this time – instead of making product choices – participants had to taste two actual food samples (carrot and cheese) and evaluate their hedonic pleasantness and indicate the more specific emotions the taste elicited (see Spinelli et al., 2014); the latter contained both positive-negative, as well as private-collective emotions (see Luomala, Sirieix, & Tahir, 2009; Onwezen, 2015). Participants were also requested to indicate the intensity of their purchase intentions toward the foods they tasted.

A 1–7 scale was always used. The served foods were reported to be organically or conventionally produced carrots (in grated form) and luxurious vs. mundane

cheese (in chunks). In reality, the food samples were always prepared using the same food product material (conventionally produced carrot and mundane cheese). The experiments were carried out in quiet classrooms equipped with three-walled cubicles.

This time, the social visibility of the situation was manipulated by leading the study participants assigned to the public condition (n=137) to believe that they were supposed to share their food responses with the researchers at the end of the experiment (this instruction was given both orally and via text in the questionnaire). Under the private condition (n=120), no such instructions were voiced. This time, the post-study interviews did not reveal any suspiciousness. The procedure took an average of 20–25 minutes. More detailed descriptions of the procedure (e.g., the entire cover story and how the foods were served) can be found in Article 1.

3.3.3 Procedure for Experiment 3: Article 2

Two separate datasets were collected using identical intercept methods in an urban and in a rural area; the studies were presented as being about “impression formation concerning food consumers.” The first (Dataset 4) consisted of 84 male mall shoppers (mean age 36.8 years, 42.9% highly educated), while the second (Dataset 5) contained 69 male food fair attendees (mean age, 42.4 years, 26.1%

highly educated). Participants were randomly assigned to one of two conditions.

Specifically, they had to evaluate (based on a picture in the questionnaire) 1) a male wearing an organic-labeled white shirt or 2) a male wearing a blank white T-shirt in terms of certain socially valued traits (e.g., trustworthiness and affluence) on a 7-point scale. In line with the well-established approach to revealing implicit effects (e.g., Brasel & Gips, 2011; Park & John, 2014), the subtle logo (a nationally well-known ladybug) served as a cue suggestive of a favorer of organic foods.

Participants were then asked to indicate how much money they would donate to this man (a freely chosen amount in Euros) if he was collecting donations for the disaster relief work of the local Red Cross. This question served to measure how (un)favorably the pro-organic male signaler was treated by other males (cf.

Nelissen & Meijers, 2011). At the end of these studies, the extent to which the study participants perceived organic food to be cheaper or more expensive than conventionally produced food across product categories was determined on a 5-point scale. The procedure took an average of 3–6 minutes (background characteristics were recorded at the beginning of the studies). A separate pre-test was conducted in the settings of both studies to ensure that the subtle organic logo used was recognizable and its meaning was understood as intended. This was

indeed the case. More detailed descriptions of the procedure (e.g., how the pre-test was carried out and the used photographs) can be found in Article 2.

3.3.4 Procedure for Experiment 4: Article 3

A total of 168 consumers (mean age 47.6 years, 49% men) participated in the study online. The study was presented as “an academic consumer impression survey.”

Participants were recruited from the panel of a Finnish market research company;

they all came from the Helsinki metropolitan area. The online platform utilized was pre-tested by the author; it did not allow a participant, for instance, to move forward through the survey before answering all the questions on a given page (nor was it possible to go back to a previous page).

Consumer panelists received an e-mail requesting that they participate in the study, in which they were randomly assigned to one of two conditions. Specifically, they had to form an image of a consumer who was presented as a regular user of either 1) organic food or 2) conventional food in terms of several socially valued characteristics (e.g., altruism and sophistication) using a 9-point scale. Food product pictures (organically vs. conventionally produced ketchup, butter, yogurt, and bacon) were shown on a screen one by one, in line with the user imagery approach (see Liu et al., 2012; Parker, 2009). Participants were also asked to evaluate users of certain conventionally produced filler products.

After the consumer image questions, participants’ attitudes toward all the studied products (scale: 1–5) were investigated. Then, their value priorities were unraveled. Schwartz’s short ten-item measure (see Lindeman & Verkasalo, 2005) was used to determine the value priorities. To be more precise, in the case of each main value cluster (power, achievement, universalism, benevolence, stimulation, self-direction, hedonism, security, conformity, and tradition), participants were asked to indicate how important that cluster was to him or her on a 7-point scale.

The author of this work is well aware that other conceptualizations can also be used to capture consumers’ value priorities; Rokeach’s (1973) and Hofstede’s (1983) value theories provide widely applied examples of such tools. However, Schwartz’s theory of values was selected because it is not only often used in the organic food realm but also applied among people with distinct cultural backgrounds (see Thøgersen et al., 2016). Finally, participants’ brand attitudes and familiarities with the products, along with their background characteristics, were investigated. More detailed descriptions of the procedure (e.g., an example of the food product pictures shown) can be found in Article 3.