LP_Surkan_Surveys_v07

Survey Workshop : 

Survey Workshop 8th Grade Project The Walker School Brian Surkan / Melissa Tulloch

Why Conduct a Survey? : 

Why Conduct a Survey? Understand a specific group / population E.g., What policies do voters care about? Better serve a target population E.g., Comments cards in the Walker Dining Hall Validate decisions, conclusions, hypotheses E.g., Do Burger King customers prefer Coke or Pepsi? Measure changes within a population E.g., Election poles, % McCain / % Obama

Target and Sample Populations : 

Target and Sample Populations Target Population (p) Define the population you would ideally like to understand? This is NOT necessarily the target audience of your research Sample Population (x) What representative portion of the target population can realistically be surveyed / reached Distribution: How should a sample be distributed? Random vs. subgroup sampling, etc. Resource limitations: What group can be reached? Self-selection risks

Validation and Profiling Questions : 

Validation and Profiling Questions Profiling questions: M/F, Age, Income, etc. Used for sorting, analyzing data Only solicit characteristics relevant to your topic Validation questions Identify respondents with clear conflicts of interest Eliminate respondents who are not taking it seriously Ask same question in different ways Ask blatant quality control questions Home Depot: In which department did you shop? Home Depot: Select #6

Validation Examples : 

Validation Examples

Subjective Questions : 

Subjective Questions Subjective Questions Quantifiable (multiple choice) / scaled questions (e.g., 1-5) Open-ended questions (Do you have any additional ideas?) Tips Order: Profiling -> Multiple Choice -> Open-ended Avoid insufficient options (e.g. Love v. Hate) Survey length (long survey = more people give up) Anonymous v. Named surveys

Delivery Options : 

Delivery Options Hand-distributed, printed copies Data entry is laborious Personal interviews Time consuming to collect AND long data entry Mail/E-mail surveys Use special subject line / e-mail alias Web surveys Online analysis ($$) / painless collection / URL posting

Web Surveys : 

Web Surveys Popular services: http://www.surveygizmo.com/ Surveymonkey.com http://www.zoomerang.com/ Double-dipping: Respondents responding twice? Set time limits for surveys to establish context

Motivating Respondents : 

Motivating Respondents Carefully Compose your respondent request Keep it short Identify yourself and your objective Explain how their responses will be used Assure them of anonymity, if applicable Prizes can Encourage non-serious respondents Make Submission Easy: Postage paid envelopes Clickable Internet link

Collection : 

Collection Collection box In neutral, safe location Hand-collection By neutral party Electronic collection Automatic and NOT entirely anonymous Collection & Transcription Service Neutral third party who keeps raw data secret

Have data, now what? : 

Have data, now what? Respondents (n): Those people who complete survey E.g., Did enough people complete the survey (often < 5%)? Filtering: Should some surveys be discarded? E.g., Surveys where the person put 1 for every answer Scoring: How to tally/weight survey answers E.g., Should 4/5 & 5/5 (very good and excellent) be grouped? Concluding: Fair conclusions based on the data E.g., If 70% of Americans plan to vote Obama, it doesn’t mean that they all like him. Perhaps 30% just hate McCain.

Survey Workshop : 

Survey Workshop 8th Grade Project The Walker School Brian Surkan / Melissa Tulloch