Survey is a common methodology in qualitative studies. According to Jamison (2006), survey serves three main purposes: to verify experimental results such as using a triangulation method; to identify a fitting hypothesis for a research study; and to relate two or more variables in a study. The advantage of using a survey include the minimization of instrumentation errors and experimental bias, consideration of the anonymity of respondents especially when dealing with very sensitive topics, and the cost effective collection of large data sets. However, the limitation of using a perception survey or attitude survey involves the reliability and accessibility of the data, the adequate representation of a data particularly in studies involving sensitive topics (i. e. LGBT community, HIV victims and others), interpretation of the findings, formulation of questions based on the perception of the stakeholders, and accountability (Herbert, 2013).
The scientific validity of a survey regardless if the sampling design for instance uses a stratified random sampling method still depends on the response rates of the study population. Web-based survey tools has allowed research investigators to expand and maximized their response rate by providing alternatives to paper-based survey methods. This type of survey tool uses the e-mails of each potential respondent to send invitations for participations and once the invitation is confirmed respondents will receive a link to a Web-based survey form (Kroth et al., 2009). Although Web-based surveys are very helpful in optimizing data collection for surveys, Kroth and colleagues (2009) still noted the importance of a paper-based survey as an auxiliary for Web-based surveys. Such use of a mix-model strategy (i. e. combining Web-based and paper-based surveys) allows the enhancement of a response rate at least in the clinical setting (Kroth et al., 2009).
According to Herbert (2003), the success and reliability of a survey study depend on how the methodological issues are addressed. Several factors could affect the data of a perception and attitude survey conducted in high school students for instance. Some of these factors include timing (i. e. if surveys are conducted during the spare time of students to unwind), sequencing and phrasing of questions (i. e. question priming), perception of stakeholders regarding the survey instruments (i. e. designing questionnaires based on guide discussions with local stakeholders), and conducting a pilot test so that the questions could be edited if they are vague, irrelevant, or very sensitive.
The methods applied by Cork and colleagues (1998) in developing and validating an instrument to measure physicians attitudes regarding computers may be modified and adopted in developing an attitude or perception study of high school graduates from public schools. Cork and colleagues emphasized that the study on the attitudes, expectations, demands, knowledge and experiences of a study population should address the psychometric properties of the variables. In developing the questionnaire for an attitude study of high school graduates of public schools within the inner city, the attributes of each high school student and the variables that they are associated with should be considered. The objective is to make a questionnaire that could produce results that could be generalized. The psychometric properties of high school graduates and their associated variables should address the following: how the scale that covers the dimension of each attribute coincides with the hypothesis, the items that are irrelevant to the set of attributes, reliability of the scales used per item set; and the extent of validity of the scale as a measure of the associated variable.
Cork, R. D., Detmer, W. M., Friedman, & C. P. (1998). Development and initial validation of an instrument to measure physicians’ use of, knowledge about, and attitudes toward computers. Journal of American Medical Information Association, 5, 164-176.
Herbert, S. (2013). Perception surveys in fragile and conflict-affected states (GSDRC Helpdesk Research Report 910). Birmingham, UK: GSDRC, University of Birmingham.
Jamison, J. (2006). Research methods in psychology for high school students. Lincoln, NE: Jennie Brooks Jamison.
Roth, P. J., McPherson, L., Leverence, R., Pace, W., Daniels, E., Rhyne, R. L., et al. (2009). Combining Web-based and mail surveys improves response rates: A PBRN study from PRIME Net. Annals of Family Medicine, 7(3), 245-248.