Are you surveying the key populations that matter to your brand?
If not, you could be missing out on insight to enhance how you make decisions about your organization.
Drafting thoughtful surveys and collecting answers from the audiences that impact your brand reveals useful insight to guide your organization.
Communications professionals should keep several things in mind when creating questions to avoid biases and defective data.
Following the creation and distribution of the surveys, professionals must then analyze the information they have collected from survey respondents.
With proper analysis and comparison of the data, communicators can learn valuable insight that informs how they make decisions about communications campaigns and tactics.
To build a survey
If your organization is diving into surveys for the first time, it’s important to consider the key components of a well-crafted survey. As with any other PR measurement tactic, consider your goals for the survey. What do you expect to learn and what insight do you hope to gain from this process?
Once your organization has established its goals, you need to consider how you will design your survey. Structure and phrasing will dictate how your respondents interpret and answer questions. The quality of your data relies on how well the survey questions are written.
Surveys can include closed and/or open-ended questions. The structure and context may provoke respondents to answer closed-ended questions differently than they would answer open questions.
For example, Pew Research Center described a poll they conducted following the 2008 US Presidential election. They found people responded quite differently to the following question when it was proposed as closed-ended and open-ended: “What one issue mattered most to you in deciding how you voted for president?” Fifty-eight percent selected the economy when provided that answer in the closed-ended version of the question, while only 35 percent provided this answer in the open-ended version.
Questions should also be written without bias. Leading questions influence how people respond, contributing to data that may not reveal the true opinions and feelings of respondents. For example, people would likely respond differently to these two questions:
- Did you find the class provided you with useful information?
- Don’t you think the class provided you with lots of useful information?
The first question lacks bias, asking the respondent’s opinion on whether or not the class succeeded in conveying useful information, while the second question encourages the respondent to answer in the affirmative about the class’ usefulness.
Also important to consider is the order and number of options available per question. In particular, professionals should consider the recency and primacy effect when constructing survey questions. The recency effect is the tendency for people to recall and select the most recent option, while the primacy effect is the bias to remember and select the first items on a list.
According to Pew Research, this is especially important to consider for phone surveys, as people are more likely to respond with the selection they hear at the beginning or end of a list. Setting up survey questions to be randomized, however, ensures respondents will receive answer selections in varying orders, decreasing the chance that primacy and recency biases will affect the response data.
Once your survey has been created and distributed, your organization is left with a pile of data that must be analyzed to extract meaningful takeaways.
Your survey data may come in one of three forms: text, numbers, and categorical. Numbers, also referred to as ordinal data, comes from questions asking respondents to select a number from a scale, such as, “With 1 being strongly disagree and 5 being strongly agree, please rate the following statement: the customer service department was helpful during my return process.”
Categorical data comes from questions asking respondents to select an answer from a predefined list. For example, categorical questions may ask, “Where do you normally purchase your groceries: Shaw’s, Hannaford’s, Whole Foods, Trader Joes.”
Text data comes from open-ended questions, such as questions asking, “Please describe your experience with the free trial of our software.”
When analyzing data, especially numerical data, Excel is an excellent tool to facilitate the management. You can use Excel for calculations, such as mean and median, to contextualize the data. Add other relevant values to a table, including minimum and maximum values, to provide context to the information.
Excel also allows users to turn the data into visuals. As with any other PR measurement process, visuals more effectively communicate the data and allow professionals to extract meaning. Plot the data on a graph, comparing it against historical data on the same topic, if this information is available, to show trends and changes.
For categorical and textual data, Excel can still be used to analyze the information. Begin by reviewing and grouping similar answers. For example, if you ask respondents to describe their experience with a customer service representative, group answers based on whether the experience described was positive, negative, or neutral. Categorical answers, on the other hand, can be grouped based on how many respondents selected each of the provided answers.
Once you have a breakdown of this data, the groupings you create can be plotted on a graph to reveal trends in your respondents’ answers. For example, if your data shows an overwhelming number of negative responses to a question about their experience with your customer service team, it would suggest that you need to dig deeper into the data to understand the source of the problem. The insight provided by this data allows you to fix the problem, increasing customer satisfaction and improving the relationship between your brand and its audience.
Even for communicators without experience in surveys, conducting this type of primary research on targeted audiences reveals insights that allows your organization to make data-driven business and communications decisions.