Skip to main content

Survey + Interview Best Practices

By February 5, 2017February 23rd, 2023No Comments

Survey Creation:

Surveys are an effective way to collect large amounts of data from a specific audience segment.  When well-constructed, surveys primarily collect quantitative data rather than qualitative data.  At Wilderness Agency, we use Qualtrics as our survey builder platform.  For an overview of how to utilize the full capabilities of Qualtrics, click here.  You should consider using a survey if your project scope meets the following requirements.

  • You need to collect a large amount of data efficiently.
  • You need to better understand your target markets.
  • You need to test a new concept or gather valuable insights into categories such as product development, consumer behavior, or logo concepts.
  • You need data to back up your strategic decision-making.

Sampling Methodologies

Convenience Sampling is when you sample the most easily accessible resources, such as distributing your survey to friends, family, co-workers, or on social media.  This method requires the least amount of effort to create a sample for your survey.  However, it is the least accurate form of sampling given that people who are surveyed may or may not be in the target population.  This form of sampling is still more accurate than guessing.

Judgment Sampling requires the researcher to use his or her judgment to determine a sample that is likely to represent the target population.  This form is more accurate than convenience sampling, but still poses barriers given that it relies on the researcher’s judgment.  An example of judgment sampling includes distributing surveys on sourcing platforms such as MTurk, where the researcher restricts the demographic or psychographic parameters of survey respondents.  While judgment sampling can be effective, it requires understanding your target population prior to distributing the survey.

Random Sampling is the most effective and accurate form of survey distribution.  With this methodology, respondents are chosen at random within the target population.  Each individual has an equal opportunity of being selected and fully represents the population.  This method of sampling is most difficult to effectively implement as it is typically time-consuming and costly.

There are sample size calculators that generate the appropriate number of responses depending on the anticipated size of your target population.  The biggest hurdle in survey distribution and collection is motivating respondents to complete the survey.  The most effective way to motivate individuals is through monetary incentives, such as gift cards, mailed rewards, and more.  Additionally, offering individuals a portion of the reward upfront can create a feeling of obligation to encourage higher response rates.

When it comes to survey errors, there are two main forms.  Sampling errors occur when the sample generated for the survey is not accurate, or reflective of the desired population.  This form of error also includes insufficient sample sizes.  Survey errors happen when there are errors in the writing of the survey, which can lead to inaccurate responses or respondents submitting random answers.  Random answering can be combatted by adding a “check” question.

Survey Introduction Tips

Survey introductions matter.  This is when you can capture the respondent’s attention.  Additionally, be brief and create a clear understanding of what you are asking.  There are three crucial items to include in any introduction:

  1.  The duration of the survey.
  2.  Why they should take the survey or how much will be compensated?
  3.  What the survey is about?

Survey Question Types

When creating a survey, there are different forms of questions that are applicable in certain situations depending on the desired data.  Listed below are the different types of questions, as well as the pros and cons of each question format.

Here are five key concepts to keep in mind when creating effective survey questions:
  1.  Everyone interprets the question the same way.
  2.  Everyone can find an answer that matches their opinion.
  3.  The questions are quick and efficient to answer.
  4.  Always provide an out response.  For example, “not sure” or “I don’t know.”
  5.  Limit open-ended questions, as this minimizes the amount of quantitative data and may encourage respondents to drop off mid-completion.

Recall Question: Respondents are asked about top-of-mind recall regarding brand, product, or service awareness through a short-answer, open-ended format.  While such questions offer insight regarding potential direct competitors, it also requires effort and may deter some responses as a result.

Consideration Question: Respondents are asked about top-of-mind recall regarding brand, product, or service awareness when provided with a list of existing competitors or other brands.  While it requires little effort from respondents, awareness does not necessarily equate with purchasing behavior.  Individuals may also be more likely to over-select.

Bipolar Scale: Respondents are asked about favorability or lack of favorability regarding a brand, product, or service on a weighted scale from negative to positive in the form of a question.  Always be sure to incorporate an odd number of selections to offer a neutral option.  When the scale is utilized correctly, the standard format of the questions allows for accurate responses.

Likert Scale: Respondents are asked about favorability or lack of favorability regarding a brand, product, or service on a weighted scale from negative to positive in the form of a statement.  Always be sure to incorporate an odd number to offer a neutral option.  When the scale is utilized correctly, it allows for fast, accurate responses.

Ranking Question: Respondents are asked to rank a brand, product, or service in relation to the other options presented.  This form of question best assesses how customers perceive your brand, product, or service compared to competitors, but does not necessarily indicate the extent of favorability.  For example, an individual ranking Apple over Samsung does not mean that the respondent likes Apple; they simply prefer Apple over Samsung.  When utilized correctly, the questions allow fast, accurate responses.  It is a common way to assess attitudes.

One-on-One Interviews

One-on-one interviews are an excellent way to better understand feelings, emotions, and rationale – essentially, interviews help to answer the “why” rather than the “what.”  Well-conducted interviews collect qualitative data rather than quantitative data.  Before you create the interview structure, it is helpful to outline the research questions and goals for the interviews.  For example, are you trying to understand customer behavior, or are you analyzing the difference between external and internal perceptions of a company’s brand?

Consider using interviews if your project scope meets the following requirements.

  • You need to truly understand your stakeholders’ or customers’ needs and motivations.
  • You need to have the flexibility to ask specific, customizable questions without worrying about survey bias or error.
  • You have a limited sample size and would value a great amount of qualitative data versus limited quantitative data with a smaller sample size.

Tips & Techniques

Try to make the interviewee as comfortable as possible.  It generally helps to structure interviews such that the interviewee starts in familiar territory, and gradually approaches more complex topics or questions.  This helps to build rapport, gain trust, and generate more meaningful qualitative data.  Remember that the goal is to understand the interviewee’s motivations and feelings; always offer ample silent “space” in the conversation.  Here are four overarching tips for successful interviews:

  1.  Ask the interviewee to elaborate with a story.  By asking them to go into further depth, you are provided with great detail and insight.
  2. Laddering, or asking “why” multiple times in different ways to get in-depth information.  This helps to uncover the underlying emotional benefits of a service or product rather than the functional or technical benefit.
  3. Elimination questions help to gauge how important or unimportant a product or service is if the interviewee did not have it.
  4. Tradeoff questions assess the value and importance of a product or service in relation to a competitor’s or a replacement.