Many research projects require the collection of new data. However, this is not always necessary - you should investigate what datasets are already available. A list of some important datasets by country can be found here.
Data collection methods can be divided into two broad categories, obtrusive and unobtrusive data collection. This refers to whether the researcher influences the units of observation while collecting data or not. Both obtrusive and unobtrusive data collection methods can be used to collect various different types of data. An important distinction in terms of types of data is between quantitative qualitative data. When data is recorded in numerical form and entered into a data matrix, it is known as quantitative data. When it is not recorded in numerical form, it is known as qualitative data.
If the target population is a set of individuals, data are typically collected using surveys or interviews. Interviews and surveys are ‘obtrusive’ data collection methods, because the individuals being studied are aware that they are being studied, and this may affect their behaviour (answers).
There are various types of surveys such as internet surveys, telephone surveys and postal surveys. Programmes such as Google Docs can help you to make an online survey, although other services are available too, like SurveyMonkey. Students and employees of the University of Twente can get access to LimeSurvey via Datalab. This allows you to collect data via the web. Responses from a survey can then be entered into a data matrix for analysis.
If only a small number of individuals are asked about a lot of topics, you normally use interviews. In ‘structured interviews’, each interviewee is asked the same questions in the same order, and usually there is a set of multiple choice answers specified in advance. Data from structured interviews can be treated much like a survey, and the results entered into a data matrix. ‘Unstructured interviews’ are used to gain an understanding of a topic or phenomenon: here, the questions often change from one interview to the next, and the interviewee is not given fixed set of answers to choose from. Information gained in unstructured interviews is not used to produce a data matrix for statistical analysis.
It is also possible to gather information in such a way that that the individuals (or companies, groups etc) that you are studying are not aware that you are doing so, and your research will not affect their behaviour. Many forms of information exist that are recorded for purposes other than scientific research, yet can be later used for this purpose. Examples include data on crime, hospital admissions, house prices, employment, or road accidents. Many of these types of data can be found on national statistics websites such as Statistics Netherlands. There are also many sources of information in the form of recorded communication that can be used as a source of data, such as newspaper coverage of particular topics over time or across countries; election manifestos of political parties etc. Content analysis is used to collect data from documentary sources. Content analysis is often used in a way that produces quantitative information that can be placed in a data matrix. A list of software available for content analysis can be found here.
- Babbie, Earl (2004). The Practice of Social Research (12th edition). Belmont: Wadsworth/Thomson. Chapters 8-11.
- De Veaux, Richard, Paul Velleman and David Bock (2008). Stats: Data and Models (2nd edition). London: Pearson/Addison Wesley. Part III.
- Becker, Howard S (1998). Tricks of the Trade: how to think about your research while you’re doing it. Chicago: University of Chicago Press.
- Hammersley, Martyn (1992). What’s wrong with Ethnography? London: Routledge.
- Yin, Robert K (1994). Case Study Research: Design and Methods (2nd edition). Thousand Oaks, California: Sage.
- Webb, Eugene J., Donald T. Cambell, Richard D. Schwartz and Lee Sechrest (2000). Unobtrusive Measures (2nd edition). Thousand Oaks, California.
- Groves, Robert M., Floyd J. Fowler and Mick P. Couper (2004) Survey methodology. Hoboken, Wiley-Interscience.