ABSTRACT

A survey of the extent to which cost, taste and health considerations impact food purchasing practices in Australia was conducted. Data were gathered from a national computer assisted telephone survey of 1109 randomly sampled householders and analysed using multiple logistic regression analysis. 88% of respondents considered the taste of food before its price, with females and people on higher incomes more likely to do so. 52% of respondents said that they considered the price of food before its health and nutritional benefits, with males, younger people and people with lower educational qualifications more likely to do so. 49% said that they purchase organic food, with people with 1 child, full-time employed and people never married more likely to do so. Overall, gender, income, education, work status, age and family size are all important predictors of food purchasing practices in Australia.

1. Introduction

This paper focuses on the extent to which cost, taste, health and ethical considerations impact food purchasing practices (referred to as FPP throughout this paper) in Australia. The specific FPP considered in this paper are: considering food prices before their health or nutritional qualities when purchasing food (i.e. price-before-health); purchasing organic foods (may be related to health and ethical issues); and purchasing foods which are more expensive because they taste better (i.e. taste-beforeprice). The paper provides data from a national survey in Australia about both the prevalence and social determinants of these FPP. By social determinants, we mean the relative effects of factors such as gender, age, income, social class, household size and composition, and educational attainment on FPP. Our paper is timely given the rising food costs, increased focus upon prevention of chronic disease through adoption of healthy diet, and the ethical and environmental considerations in relation to food production and transport. Thus the need to understand changing consumer food-related practices has moved to centre stage in public health policy. In learning the details of everyday dynamics of FPP, this paper will also shed light on the “complex range of factors which operate to produce and/or sustain “unhealthy” lifestyles and (by extension) overweight and obesity” [1].

FPP are embedded in the everyday details of life— what French anthropologist Bourdieu [2,3] refers to as the habitus, which is shaped by sociocultural and economic settings. Food and food practices are not only symbolic of, but are deeply engrained bodily performances of identity, class and social relations, and are transmitted, learned and reproduced in families and across generations. These bodily ways of being and knowing are largely unconscious and taken-for-granted. However, Sayer undertakes a critical analysis of habitus, arguing that habitus is more flexible and malleable than Bourdieu originally asserted [4]. Indeed, Sayer argues that it is difficult to see how “resistance” would be possible if individual identity are largely a product (both complying with, and complicit in the production of their habitus) of their social and cultural milieu. Bourdieu [2] argues that each individual is born into particular cultural and class meaning systems that code the body in ways of “standing, speaking and thereby of feeling and thinking” (p. 32). Therefore, the “food practices” within this paper are not simply individual “behaviours”, but represent an interrelationship between the individual and the wider classed and gendered structures within society.

2. Consideration of Food Costs as a FPP

Cost has been identified as a major consideration in food choice [4-7]. As in many other countries, Australian consumers have recently had to accommodate increases in costs of basic food [8]. During the financial year 2007-8, overall food prices rose 3.9%, while some basic food prices rose more sharply: cheese by 14.2%, milk by 12.1%, poultry by 11.0% and bread by 6.8% [9]. Food cost plays a significant role in mediating food choice among low socio-economic status (SES) groups [10,11], who often have to cut back on food spending to make room for other essentials such as housing and utilities [12-15], leading to decreased food security [16].

Food insecurity is associated with obesity [17,18] and obesity related disease [19,20]. These elevated rates of obesity among the food insecure is thought to result principally from increased consumption of foods high in fat and or sugar that are typically cheaper, more available, heavily marketed and simpler to prepare than healthy alternatives [6,7]. Furthermore, the health consequences of food insecurity go beyond obesity and include nutrient inadequacy [21], self reported health [20] and compromised child health [17]. Data collected in South Australia estimates the food insecurity rate to be approximately 7% [22]. However, this increases among at-risk groups including: unemployed (11.3%), rental households (15.8%) [23], those identifying as Aboriginal or Torres Strait Islander (23%) [24] and recently arrived refugees (71%) [25].

There are a number of recognised social determinants of food insecurity, such as the unaffordability of healthy food for lower SES groups [26,27], rising food prices in Australia [11], higher food prices and greater density of unhealthy food options in socially disadvantaged areas [28,29], employment status [30], educational attainment [31] and access to private transport [32]. We assess a number of these variables, in addition to others, when examining the nature and extent for FPP in Australia.

3. Purchasing Organic Food as a FPP

The market for organically produced foods is high within the developed world. In Australia, the organic food market increased to $947 Million in 2009, with sales up 50% from the previous two years [33]. In the UK, organic food sales were over ₤2 Billion in 2007, but reduced to ₤1.8 Billion by 2009, possibly due to the economic downturn [34]. Organic food tends to be more expensive than non-organic food, and evidence suggests that consumers are willing to pay the higher price for organic foods based on their perceived health, nutritional and taste benefits [35]. Indeed, a number of studies have found that consumers perceive organic food to be both healthier and of higher nutritional quality than nonorganic food [36-38]. Of particular concern to consumers is the use of pesticides. Canadian research has found that women, people aged 18 to 24 years and from larger households are more concerned with regulation of pesticide use [39]. Conversely, Australian research found significantly less concern with pesticide use among 18 to 24 year olds than older participants [40]. Two recent systematic reviews found however that there is no evidence of nutrition-related health benefits resulting from the consumption of organic food in comparison to nonorganic food [41,42]. These systematic reviews did not undertake meta-analyses due to methodological diversity of studies examined and did not examine the public health or environmental benefits of organic food. Nevertheless, the literature suggests that consumer perceptions about the health benefits of organic food, often promulgated by organic food companies, leading to increased willingness to pay for organic products may be unfounded.

4. Taste and Food Choice

A final consideration is taste. Taste has been identified as being a significant contributor to food choice [5], particularly for younger people who have less immediate concerns with health [43]. Cultural and gender differences have been noted in the relative importance placed upon taste and health. Participants from countries such as the US [44] and UK [45] place greater importance upon health concerns and less upon the pleasure of eating than participants in countries such as France, Belgium and Finland [44,45]. Likewise, women generally place less concern upon the pleasure of eating than men also demonstrating greater concern with the healthiness of food [44,45].

This paper provides data from a national survey in Australia on the prevalence and socio-demographic predictors of three FPP: consideration of price-beforehealth; purchasing organic food; and consideration of taste-before-price.

5. Method

This study was primarily concerned with identifying the nature and level of consumer trust in the Australian food supply [45-48], however our national survey also provided a unique opportunity to assess the prevalence and social determinants of FPP.

Households in Australia with a telephone connected and the telephone number listed in the Australian electronic white pages were eligible for random selection in the sample for this study. All selected households were sent an approach letter on Flinders University letterhead which detailed the purpose of the study and advised that the household would be receiving a phone call for an interview. The person, aged 18 years or over, who was last to have a birthday, was randomly selected within each contacted household to complete the survey.

In order to test question formats and sequence, and to test survey procedures, a pilot study of 52 randomly selected households was conducted prior to the main survey. Information obtained from the pilot was used to improve the questionnaire if needed. 

Professional interviewers from a contracted agency conducted the study using Computer Assisted Telephone Interview (CATI) methodology from October to December 2009. A minimum of 10 call-backs were made to telephone numbers selected, to interview household members and different times of the day or evening were scheduled for each call-back. Non-contactable or responding persons were not replaced with other respondents. Each interview took an average of 14.5 minutes to complete, and ten percent of each interviewer’s work was validated by the interviewer’s supervisor for quality purposes.

Of the initial sample of 4100, a sample loss of 1408 occurred due to non-connected numbers (1060), nonresidential numbers (135), ineligible household (139) and fax/modem connections (74), leaving 2692 phone numbers eligible for survey phone calls. After refusals, terminated interviews, non-contactable households, deaths, unavailable respondents and respondents who did not speak English, 1109 interviews were completed. This generated an overall sample response rate of 41.2%.

To address the issue of assessing FPP, three survey items were examined, all of which were framed as follows: “I will now ask you to consider your food purchasing habits in general. This includes green grocers, sandwiches, restaurants, as well as other take away outlets.” The specific survey questions were:

• Do you buy products that are a bit more expensive if the taste is better?

• Do you buy organic foods?

• Do you consider food prices before health and nutritional qualities?

Respondents were provided with response options assessing the frequency of the aforementioned items, ranging from “Often”, “Sometimes” and “Seldom” to “Never”. “Don’t know” responses as well as refusals to answer a particular question were recorded as further response options.

One of the obvious limitations of the price-beforehealth and taste-before-price variables are that we do now know how much “a bit more” is in relation buying tastier food and we do not know exactly how much people are willing to consider when thinking about the health and nutritional qualities of food. However, the purpose of this paper is not to provide an econometric analysis or contingent valuation of food vis a vis health or taste, but to paint a picture of the types of people more likely to consider price-before-health, taste-before-price and purchasing organic foods in Australia.

Data Analysis

Demographic variables included in the analyses were age, sex, number of people in the household, number of children under 18 years of age in the household, marital status, work status, education, annual household income, the Socio-Economic Indexes for Areas (SEIFA) Index of Relative Socio-Economic Disadvantage (IRSD) as well as the Accessibility/Remoteness Index of Australia (ARIA). However, only the statistically significant predictor variables are presented.

Statistical analyses were carried out using SPSS version 17.0. As samples such as these may be disproportionate with respect to the population of interest, weighting was used to compensate for differential nonresponse and correct unequal sample inclusion probabilities. In order to reflect the Australian population structure 18 years and over, the data were weighted by age and sex reflecting the Australian Bureau of Statistics 2007 Estimated Residential Population.

For the FPP outcome variables (i.e. the three items addressing the prevalence of FPP), dichotomization procedures were also applied: “Often” and “Sometimes” responses were added together to create one level of the outcome variable (“Often/Sometimes”), while responses indicating the respondent to perform a particular action “seldom” or “never” were combined to generate the second outcome level. Responses in the form of “Don’t know” and refusals to answer a question were not included in the present analysis.

All demographic predictor variables were entered into the analysis as categorical variables, the individual levels of which are summarized in Table 1. Bivariate logistic regression analyses were performed to examine the relationship between the individual demographic predictors and the various food purchasing habits. Only items showing an association at the p < 0.25 level were entered into multiple binary logistic regression analyses [49]. Following suggestions by Field [50], for the purposes of the present investigation the method of choice for conducting regression analyses was to enter relevant predictor variables in one block rather than stepwise procedures. Predictor variables that were entered into the model but returned as not significant were in turn tested against

models containing only significant predictor variables. This process allowed for the comparison of several models, resulting in a final model containing only variables, which significantly contributed to the model fit. For each outcome variable, predictor variables included in the regression model were checked for multicollinearity.

6. Results

Table 2 presents descriptive analyses of the responses to each of the three FPP. Of the respondents 88% said that they “Sometimes/Often” pay more for food products if the taste is better (n = 968) compared to 12% (n =130) who said they “Never/Seldom” pay more if the taste is better. Responses were more evenly distributed for the survey item addressing the purchase of organic products (53% (n = 575) “Never/Seldom” versus 47% (n = 513) “Sometimes/Often”) and whether participants considered food prices before health and nutritional qualities (48% (n = 522) “Never/Seldom” versus 52% (n = 573) “Sometimes/Often”).