Planning a Successful Survey

I could use some guidance. I recently graduated from college and started a new job at a local non-profit. The organization is committed to economic development in low-income communities. I was a volunteer with the organization throughout college, so it made perfect sense to continue working there after graduation. This past week, however, I was given the responsibility of planning the data collection for our annual survey. The survey is the one of the biggest projects we have, which means this is a huge deal. The only problem is that I don’t know too much about survey planning and/or data collection. I feel like I’m way out of my league. Any insights that might aid me would be much appreciated.

It already sounds like you’re well on your way to a fulfilling career, especially because it combines a righteous cause with both intellectual and moral rigor. There’s no shortage of disadvantaged peoples and exploited communities in dire need of compassionate help. As recently as 2013, researchers at The World Bank estimated that 10.7 percent of the global population lived on less than $1.90 a day. The Center for Poverty Research at the University of California, Davis, estimates the domestic poverty rate at 12.7 percent — higher than the global average, despite being one of the wealthiest super-powers.

Aside from being relevant to your professional career path, those statistics rely heavily (if not exclusively) on conducting successful surveys. Fortunately, that means there’s already plenty of credible resources available to help you plan a survey of your own. One such example is the Harvard University Program on Survey Research, which published a highly-informative survey tip sheet made for beginners and veterans alike.

Consulting that tip sheet on a regular basis would be a very sound strategy. The faculty and graduate students at the Program on Survey Research maintain countless resources that might aid you. There you can find information about survey modes (e.g., phone interviews, online questionnaires, in-person conversations, etc.) and how to design questions that reveal unbiased answers from respondents. You can learn about question ordering, reference frames, and how to avoid unnecessary complexity. Much of what you’ll read should be intuitive enough to implement with some concerted practice and repetition.

For a real beginner’s guide replete with visual examples, staff writers at Science Buddies® have explained how to design a survey from scratch. What they don’t do, unfortunately, is explain how you can interpret the results of the survey. That’s when techniques such as cross tabulation and regression analysis become essential. Cross tabulation is best used to compare mutually exclusive categorical data. Eye color is one example of categorical data. Regression analysis, on the other hand, is best used to determine the impact of certain variables. In other words, it reveals the answers to the questions: Which factors matter most? Which factors can be ignored? How do these factors interact with one another? And, of course, how certain are we about all these factors?

While crosstabs and regression are two of the most popular methods of analysis, they certainly aren’t the only ones. You should consider exploring a wide array of techniques. One final resource you might consider browsing was written by Stephanie Beadell at Zapier. She takes more of a business perspective than an academic one, but much of what she suggests essentially echoes her counterparts in academia. Cultivating a wide range of relevant knowledge should prove helpful as you make progress on your own project. That way you’ll be well-prepared no matter what survey design you choose.

“The test of a civilization is in the way that it cares for its helpless members.”  — Pearl S. Buck

Content Provided by Scholarship Media