The following module describes potential sources of data for research problems and possible methods of data collection, including sampling and survey techniques.
- To list potential sources of data and data collection methods.
- Describe types of survey techniques and factors that should be considered with each method.
There are numerous ways to collect data including experimentation, testing, and surveys. Surveys are one of the most common methods of data collection and will be the focus of this discussion. Data collection through the use of surveys may be done using one of the following four methods: mailed surveys, telephone surveys, personal interviews, or web-based surveys. Several of the resources associated with this page provide additional information on each type of survey, including the pros and cons of each. The choice of administration of the survey questions will depend on the purpose of the study, the timeline, the budget and available staffing.
Perhaps, more important than the type of survey delivery are the development of the instrument and the creation of the questions to be included. Researchers have two choices: Utilize an existing measure or create your own measure.
Regardless of which one the research chooses, development of the survey is a critical step. The quality of the data collected will only be as good as measure used to obtain the data.
Existing measures are available from a number of online sources and are described in many books including the Encyclopedia of Psychological Assessment, Test Critiques Compendium, and Measures for Clinical Practice just to name a few. Measures can also be obtained from such places as the Buros Institute of Mental Measures and the ERIC database. Finding an appropriate existing measure can also be done through literature reviews and consultation with fellow faculty and other researchers. The advantage of an existing measure is that it has been created and tested and therefore, it ensures great reliability and validity. It is also saves valuable time. Sometimes, however, researchers find it necessary to develop their own measure because there is no appropriate pre-existing measure, the financial cost is too high, or the researcher does not have the expertise to use an existing measure. There are a number of challenges associated with creating a measure, however. The researcher must create effective directions, seek guidance on how to write the survey items appropriately to ensure reliable data, and create a scoring system. While creating your own measure is more time-consuming and requires a great deal of work to be certain that it will produce the desired data set, it is a cost effective way to collect data. Once the measure itself has been determined or constructed, sample size and reducing bias in the sample will also need to be considered. For a full summary and a comprehensive discussion of all of these issues relating to the use surveys in research, view the following SlideShare presentation regarding survey research and design. This presentation covers research methodology, the research process, types of surveys, survey construction, levels of measurement, bias, and sampling.