JMIR Publications

ABSTRACT

Background:

To date, the evaluation of diet has mostly been based on questionnaires and diaries that have their limitations in terms of being time and resource intensive, and a tendency towards social desirability. Loyalty card data obtained in retailing provides timely and objective information on diet-related behaviors. In Finland, the market is highly concentrated, which provides a unique opportunity to investigate diet through grocery purchases.

Objective:

The aim of this research was to investigate and quantify the selection bias in large-scale (n=47,066) loyalty card (LoCard) data, and to correct the bias by developing weighing schemes. Secondly, we investigated how degree of loyalty relates to food purchases.

Methods:

Members of a loyalty card program from a large retailer in Finland were contacted via email and invited to take part in the study, which involved consenting to the release of their grocery purchase data for research purposes. Participants’ sociodemographic background was obtained through a web-based questionnaire and compared to the general Finnish adult population obtain via Statistics Finland. To match the distributions of sociodemographic variables, post-stratification weights were constructed by using the raking method. The degree of loyalty was self-estimated on a five-point rating scale.

Results:

Compared to the general Finnish adult population, there were more women (65% vs. 51%), individuals with higher education (57% vs. 33%), and employed individuals (61% vs. 52%) in our study sample. Individuals under 30 (14% vs. 18%) and over 70 years of age (8% vs. 18%) as well as retired individuals (21% vs. 32%) were underrepresented. The food purchases differed by the groups of degree of loyalty with a higher share of vegetables, red & processed meat, and fat spreads purchases in the highest groups of loyalty.

Conclusions:

Individuals who consented to their loyalty card data being used for research purposes tended to diverge from the general Finnish adult population. However, the high volume of the data enabled the inclusion of sociodemographically diverse subgroups and successful correction of the observed bias. In addition it seems that food purchases differ across degree of loyalty groups, which should be taken into account when researching loyalty card data. Despite its limitations, loyalty card data provide a cost-effective tool to reach large groups of people inclusion of including the hard-to-reach population subgroups. Importantly, loyalty card data enable the monitoring of longitudinal trends in grocery purchases.