By: Heidi Dietzsch

As data collection methods evolve rapidly in line with technological advances, some become obsolete while new methods emerge.

It wasn’t too long ago that researchers relied on paper questionnaires sent through the post. Now short intercept surveys (SIS) that are completed on respondents’ smartphones in real time are becoming increasingly popular.

The postal pen-and-paper questionnaire technique was laborious and time-consuming – all the hassle and red tape, combined with a massive potential for errors, was a nightmare to deal with. Not many researchers hark back to those “good old days”.

SIS, in contrast, are fast and respondents complete it while they are busy with a specific action (for example, while they are in a store, they can document their shopping experience) and their answers are immediately available for analysis. A distinct advantage of SIS is that respondents don’t need to solely rely on their memories.

Debates prevail among researchers and scholars on which data collection methods are the most effective and produce the most accurate data. There is also speculation that different methods that include the same questions might yield different results.

Personal interviews, telephonic interviews and online surveys are the most prevalent data collection methods in market research.

One of the benefits of personal interviews is that a skilled interviewer who is able to establish rapport with respondents, can ensure a really deep conversation. However, many people might be hesitant to share sensitive information due to the non-anonymity of the setting.

During a personal interview an interviewer can motivate a respondent when a questionnaire is very long and respondent fatigue is setting in. When this is happening in an online survey, respondents are likely to skip questions or abandon the survey entirely. Interviewers are also on hand to explain difficult questions and complex rating scales which could be interpreted incorrectly when other methods are used. However, it is also likely that interviewers themselves might fill in answers inaccurately. Another problem is that when an interviewer is completing a long questionnaire with complex routing, errors are possible, especially when it is necessary to page back and forth through a cumbersome paper questionnaire.

When telephonic interviews are conducted, computer-assisted telephone interviewing (CATI) software is usually used which diminishes human error to a large extent. However, CATI call centres can be highly pressurised environments and mistakes can thus still occur. For example, it is quite easy to complete a rating scale question the wrong way around. So, when an interviewer ticks “very poor” when a respondent means “excellent, or “agree” when a respondent says “disagree”, it can render data useless.

A telephonic interviewer’s skills and professionalism can also contribute towards the success of an interview and the quality of the data. Respondents are much more likely to agree to an interview if an interviewer requests one in a calm, friendly and polite tone. If this is the case, respondents are also more inclined to answer questions honestly and patiently, while sticking it out to the end. Of course, the opposite is true when interviewers are rude and incompetent.

Open-ended questions are not always ideal in telephonic interviews. Interviewers might not be able to type respondents’ full answers and therefore the gist, tone and context can get lost. However, as opposed to face-to-face interviews, telephonic interviews are more conducive for the sharing of information that might be regarded as personal, due to the less intimate setting.

Today many researchers are using online surveys to collect data. They are deemed faster and less expensive and can also reach a large audience and are also, some argue, more reliable.

A great advantage of online surveys is that they allow respondents to complete them at their own pace at a time that is convenient to them. Respondents can thus read all questions carefully and make sure they understand them – this will naturally lead to good quality data.

However, the opposite can also occur. When respondents receive an invitation to participate in a survey during a busy workday, they might complete it hastily without giving it the necessary consideration and attention. Data quality might suffer as a result. Also, without the guidance of an interviewer, the possibly exists that respondents might interpret rating scales and questions incorrectly. This will not usually be a problem for people who are fluent in using the internet though. In the same vein, when a question requires knowledge about a certain subject, the data resulting from online surveys are usually of a better quality than the data collected through other modes.

Online surveys are a great method for the gathering of qualitative data. Respondents can express themselves clearly and in their own words, without the possibility that an interviewer might interpret their thoughts incorrectly. Also, when researchers need to conduct a study with a sensitive topic that requires respondents to divulge very personal information, online surveys are ideal. Controversial topics that might cause respondents to think twice before taking part include personal income and finance, sexual behaviour, illegal behaviour, drug and alcohol use, religion and health – especially mental health. The total anonymity of online surveys, however, might make respondents much less hesitant to share such information.

It is important that researchers use the appropriate data collection method for the type of data they wish to collect. Equally important is that the chosen method suits the respondent base that will be used. It makes little sense to try to conduct an hour-long telephonic interview with busy business executives, while an online survey will be inappropriate for rural inhabitants who may be less tech-savvy and might not have proper internet access.