The short article below addresses some of the more elementary aspects of survey research we are asked about by our subscribers. Before delving into the topic, I will first provide a definition of survey research posted by WhatIs.com. They define survey research as “the collection of data attained by asking individuals questions in person, on paper, by phone or online.” SnapSurveys further elaborates that there are four major benefits of survey research: 1) low cost, 2) samples allow you to describe large populations, 3) they are amendable to many delivery modes, and 4) they are dependable in getting honest and reliable responses.
When researchers go to their tool bags for a preliminary and inexpensive instrument to test the waters of public opinion, consumer sentiment, market forces, and other elements of interest to the research world, the quick and inexpensive survey is one of the first tools to be employed. When professionally developed and given to the appropriate sample from the population of interest, it can both help illuminate the research question the survey seeks to answer but also the response data can provide a more focused direction for further research or additional surveys. Surveys have indeed helped organizations of all types better serve their customers and reach new customers. Surveys have provided information to researchers that have been the impetus for new products and services and in some cases new companies. It is a wonderful instrument for finding out what people and companies need while also alerting the researchers as to what is not needed. This paper will focus on online and paper surveys since they are the most used survey instruments today. We will look at the basic framework that is important in survey research.
Survey data quality is an essential element of a good survey. Safeguards are put in place by survey researchers to ensure that problems that may compromise data quality are minimized or avoided. Throughout the survey, the survey researcher provides quality controls through constant monitoring of the process, analysis, and verification of essential aspects considered as quality indicators. The process ensures that updates to the quality assurance procedures are done to meet the working standards of quality assurance. The methodologies chosen for a research project are significant contributors to the quality of the survey and the collected data. Quality control measures vary according to the type of study undertaken. One such measure is the computerization of programs that play a significant role in controlling questionnaires, managing the sample used for data collection, and ensuring that appropriate item presentation occurs.
Survey researchers managing the survey process are trained to design and select items that provide meaningful answers to the research questions in the study. Professional training enables researchers to learn how to develop and administer the questionnaires effectively in order to achieve high quality control; survey researchers are required to be trained to ensure that the best quality of data is collected through the questionnaires (Lavrakas, 2008). The item development process is essential in a survey, and when the researchers are trained in appropriate writing and pilot testing of questionnaire items, they can fine tune those items to provide the most useful information. In line with this, the questionnaires should be drafted in such a way that the respondents provide essential information. All the major aspects of the research question(s) should be covered in the survey(s). Ideally, the survey is computerized so that the management and monitoring process is standardized.
The context and wording of questionnaire items must be carefully examined. This is also true for the directions. Even the presentation order of items must be examined since it can influence results. Normally, the questions should be framed in such a way that the emotions of the respondents are not overly aroused. Maintaining a standard emotional context among the respondents will usually lend to the validity of the survey research. Of course, there are times when you want the survey item to elicit an emotional response. This is frequently the case in personality type surveys. The selection and management of samples should be done with a sampling methodology that will ensure that the respondents are representative of the population from which they are drawn. The data collected from such a process will minimize error and be more valid and reliable. This enables researchers to make more accurate characterizations of the population from which the sample was drawn. In making sure that proper samples are drawn, it is essential to understand the characteristics of the population, including the various subgroups within the population. The samples can then be selected to reflect all the different sub-groups, and this helps prevent any bias that may cause the responses to overrepresent or underrepresent a sub-group in the overall responses. This allows the researcher to not only provide survey information for the population but also for the different subpopulations. Sample design and coverage are the issues that have attracted vast attention (Lavrakas, 2008). A sample can be drawn from a target population, but poor representation can occur when the critical sample is not drawn. The quality of the data will likely be compromised.
When it comes to the use and abuse of surveys and respondent data, it is essential to note that a questionnaire survey is one of the easier and cheaper research methods used in studies. Anyone, including those not trained in survey design, can quickly throw together a survey questionnaire to put on their internet site without regard to sampling protocol or quality development. The use and abuse of data collected via untrained individuals has resulted in much disrespect being accorded to the method. Criticism is rightly made that such surveys are not valid and reliable, and the use of collected data cannot be generalized to any population. Some of the most obvious examples of survey abuse occur before major elections. A campaign will develop a survey that deliberately includes leading items to encourage respondents to answer in a predetermined manner. Once the item responses are tabulated and summarized, they will support positions held by the campaigning party.
Another tactic used on phone surveys is to tailor one’s tone of voice to encourage the respondent to provide a particular response. Often that response is what is considered a “politically correct” response. In a one-on-one phone survey, the respondent may be hesitant to reveal their true feeling in answering a question if they feel it is not politically correct. Survey companies today are more cognizant of this phenomenon and have taken measures to minimize such response errors. Ensuring anonymity via an online survey can help prevent respondents from making a politically correct response if it is contrary to their belief system. While this is a strength of the online system, online surveys frequently have low response rates which jeopardize the representativeness of the results. To mitigate the low response, increase the use, and reduce the abuse of data from survey techniques, survey researchers can implement pre-notification of the respondents through emails or postcards before the research is conducted. Follow-ups can also be done to ensure that there is accuracy in the information given by the respondents because there are cases where respondents guess responses when in doubt or rushed to complete. Survey research intends to acquire data that is representative of the target population. Any limitations in the survey and its results must be detailed in the summary of results with appropriate cautions on misuse of the data.
In many cases, readers tend to opt for surveys that contain familiar content so that they are knowledgeable of their responses. Importantly, respondents do not want to be limited to the choices of the researcher because they may have a different opinion regarding the topic. Therefore, it is important to pretest survey items and to provide enough options from which to select. Personally, I have found it useful to provide open-ended responses to survey items where respondents can put in their own response should they not want to select one of the options on the survey. This can provide very useful information to the researcher not only about the item integrity but also about an area or response not previously considered by the researcher.
Despite some of the limitations inherent in surveys, they are a critical element of many research designs. While in and of themselves, surveys may not provide the definitive answers a researcher seeks, they frequently point the direction in which research should proceed. In many cases, they provide marketers enough information to develop effective marketing campaigns. If you need information in a hurry, surveys are a great instrument to provide that information quickly and inexpensively. Some information is better than no information. Another consideration is that business and marketing dynamics are changing so quickly today that by the time longer research projects complete, the information obtained may no longer be relevant.
As mentioned previously, cost is one of the aspects where surveys distinguish themselves for researchers (Eaden et al. 1999). In line with the changing dynamics of today’s society, the distribution and management of questionnaires can be done online for the convenience of both the researcher and respondent. Unlike many other research techniques, online survey research is easy to employ, and no paperwork is needed that can make the process costly. Travelling is also another issue when it comes to the cost of research techniques. The survey does not require the researchers to travel to the respondents because the process is done online. The context of value can also be looked at in terms of the time it takes for the research to be undertaken. Information is acquired quickly and within a short timeframe. This means that the researcher can save the resources and use them for other projects; the funds include the time saved when less time is needed than required for original in-group research. The researcher can carry out several surveys at once. Several questionnaires can be constructed and sent to numerous samples. This can result in a lot of useful information without excessive spending.
Online surveys are conducted so that respondents do not have to reveal their identity. The information acquired from the techniques is professionally developed and of high quality because it is first-hand, and the participants are confident in the process because their identities are not compromised. The application of this technique is ideal for research topics that are considered sensitive in the society and do not require the identity of the respondent. More so, the participant has the liberty to give all the information they possess regarding the topic when they know that they cannot be held accountable for taking part in the research. Accuracy of data is achieved with the survey because the assurance of confidentiality makes the participants more likely to provide candid and valid responses to the study.
The data obtained with both paper and online surveys is easily collected and tabulated. For paper surveys, the data is obtained after scanning the survey documents with a high-speed scanner. Online survey responses are collected into a database. Both can be downloaded for analysis by a variety of different software products. Initial results can be obtained the same day the data is downloaded. The sample size needed for research that uses the survey technique is almost always adequate for the researcher to get the required information (Eaden et al. 1999). In many cases, researchers tend to underestimate the importance of a reasonable sample size for research. Firm conclusions about a research topic need a large sample size so that the data collected can be analyzed from the different perspectives of subpopulation participants. When a researcher or research organization uses a survey for its data collection, it can reach thousands of participants. With a large research sample size, the accuracy of the data collected is expected to be higher than the data from a small sample size. With the confidence associated with a large sample size, the researcher can quickly reach firm and sound conclusions, even when addressing the smaller subpopulations in the sample.
Online surveying is carrying out research using online platforms. They include electronic surveys where computerization of all the programs that enhance the management of surveying and data collection, web-based surveys that are conducted on organizational websites, and even email surveys. In most cases, effectiveness is achieved when the researcher or research organization obtains sufficiently large and representative samples. Website survey researchers who have large subscriber lists from which to deliver online surveys can ensure they have adequate and representative samples if the use appropriate sampling methodologies and the subscriber lists constitute the population or are representative of the population of interest.
Eaden, J., Mayberry, M. K., Mayberry, J. F (1999). Questionnaires: the use and abuse of social survey methods in medical researches. Publication of the Fellowship of Postgraduate Medicine.
Lavrakas, P. J (2008). Quality control. Encyclopedia of Survey Research Methods. Sage publishers. Available at https://dx.doi.org/10.4135/978141293947.n422