Multi-mode data collection
Mixed mode data collection in official surveys
Nino Zajc ()[1], MojcaNočRazinger ()[2]
Keywords:multi–mode, data collection, mix mode, survey designs, web mode, survey errors, individuals, business, official statistic
1.Introduction
Choosing a research plan that ensures optimal data quality and optimal use of resources is a major challenge of the modern survey research. Optimal selection is influenced by the decline in participation. More and more researchers therefore decide to use combined modes with expectation of error optimization and optimal resource usage in the survey implementation (Dillman and others 2009). The article will present the problems official statistics is facing in surveys of persons, households and businesses. We will present theoretical and practical approaches to combine different methods in all phases of the survey process. We will show errors and different approaches of combining data collection methods to improve response rates. We will discuss positive and negative sides of combinations. We will focus on the web mode and compare its use and benefits with other modes used in official surveys.
2.Methods
We will use results from pilot surveys where we used quantitative research methods. We will also use internal resource optimization (internal SURS documentation 2015) and quantitative results from the BLUE-ETS project (Blue-ets project documentation 2015).
3.Results
For the household surveys we will show how we have decreased coverage problems by offering the web mode. We will also show how offering another method (from uni mode to mix mode survey) affects the response rates and data quality. For the enterprise surveys we will show how the change in the mode influences the timing of the response, e.g. when you introduce web reporting the data are available sooner for processing. But it depends on the enterprise acceptance of web reporting, data availability in the accounting system, periodicity of the survey, etc. Enterprise reporting also depends on communication. Basic business correspondence rules should be used. It is important to use as many different modes as possible (snail mail, e-mail, telephone) and address them to different levels inside the enterprise, e.g. director, accountant, reporter. Proper communication method and timing enhance the response rate and acceptance of new modes such as web.
Figure 1.Web response rates by segment of population with and without determined telephone number.
Figure 1 shows that by offering a web questionnaire we reached a larger share of young people for whom at the integration with the phone book we could not determine telephone numbers. We also reacheda greater proportion of elderly people with telephone numbers. A larger share of employees and higher educated people without telephone numberswas also reached.
4.Conclusions
One approach to solving the problem of rising survey errors and decreasing response rates is to use combined methods of surveying (mixed mode). Such an approach combines the advantages of the different methods and at the same time reduces the disadvantages of these methods (De Leeuw and Hox 2011). We can decide to use mixed mode methods in all phases of the survey process. However, researchers’ unreasonably deciding to use mixed modes also gives rise to negative effects on data quality and survey costs.
References
[1] Blue-ets project documentation. 2015; (available on 21st October 2016).
[2] De Leeuw, E. D. and Joop J. Hox. 2011. Internet surveys as part of a mixed-mode design in Social and behaviour research and the internet, ed. Marcel Das, Peter Ester and Lars Kaczmirek, 45-72. European association of methodology.
[3] Dillman, A. Don, Jolene D. Smyth and Leah Melani Christian. 2009. Internet, Mail, and Mixed Mode Surveys: The Tailored Design Method. Hoboken, NJ: Wiley.
[4] Internal SURS documentation. 2015.
1
[1]Statistical Office of the Republic of Slovenia
[2] Statistical Office of the Republic of Slovenia