Data Needs for Consumer and Retail Firm Studies

How might in-home scanner technology be used in budget surveys?

“…In the US, for example, Homescan households appear to have fewer members on average than those in the CE (Huffman and Jensen, 2004) whilst in the UK, Kantar households have more members on average (Leicester and Oldfield, 2009). In terms of income, there is conflicting evidence in the US: Huffman and Jensen (2004) suggest Homescan households have higher incomes than those in the CE, whereas Perloff and Denbaly (2007) find that scanner data households (in data collected by IRI) have lower incomes than Census averages. In the UK, Worldpanel households were found to be poorer on average and more likely to be unemployed, or part-time workers (Leicester and Oldfield, 2009).…”

“…If a statistical agency were planning to introduce a scanner component into a random sample then this may not be a problem (other than possibly non random refusal to use the technology), but if the idea were to use existing scanner data then this might be a drawback for any statistical comparisons or imputation across surveys. Westat (2011b) and Perloff and Denbaly (2007) are both critical of commercial scanner data collectors for releasing little information on sampling methods, response rates, attrition rates and so on, and suggest caution in relying on existing scanner data for these reasons.…”

Research Note—Generating Shareable Statistical Databases for Business Value: Multiple Imputation with Multimodal Perturbation

“…While the potential value of these data is generally understood, privacy and confidentiality concerns limit its use. For example, although supermarket scanner data matched with medical records would be invaluable to health researchers, such a linked database creates serious privacy concerns (Davenport et al 2007, Perloff andDenbaly 2007). Other examples include the Netflix prize, which involved distribution of altered customer data that in some cases were later identified by researchers (Narayanan and Shmatikov 2006), and the AOL data scandal, in which that company divulged personal information, leading to class action lawsuits.…”

Evidências da sofisticação do padrão de consumo dos domicílios brasileiros: uma análise de cestas de produtos de consumo doméstico

“…As empresas varejistas têm reagido a essas alterações por meio de uma série de inovações, dentre as quais podem ser citadas: a expansão de diferentes formatos de lojas; o aumento da variedade de produtos, principalmente de marca própria e de alimentos orgânicos; além do canal on-line que busca trazer conveniência a um consumidor cada vez mais consciente e exigente (PERLOFF, DENBALY, 2007).…”

“…Para cada domicílio é designado um fator de projeção (peso), que é baseado nos dados demográficos do lar e tem como objetivo agregar os dados de domicílios com características similares, buscando com isto uma representatividade da amostra tanto no nível de mercado, da região e do país. Assim, os fatores de projeção refletem o desenho da amostra e cada fator reflete a representação de cada domicílio na população do país MUTH, 2007;PERLOFF;DENBALY, 2007). Pode-se representar o fator de projeção da seguinte forma:…”