Hypothesis testing means proving or disproving an idea, theory, or hypothesis using data from high-quality sources.

Wonder’s clients often want to know if their hypothesis is true or false so they can develop a product or service to address a problem. Knowing if their hypothesis is valid or not can help them refine their idea or change directions if necessary.

Here are some example questions:

• Is it true that most Latin American snacks in the United States come from Mexico?
• Provide or disprove that millennials will not buy products from companies that don’t have a corporate social responsibility initiative.

Minimum Requirements

• Provide 1 hypothesis per request where you will prove or disprove a theory.
• You will provide the equivalent of 2-3 insights.
  • 2-3 headers with a minimum of 3 bullet points each.
• Provide a minimum of 1 image per hypothesis.

Locating Information for Hypothesis Testing

• Use scientific papers, whitepapers, research studies, official surveys, scholarly (peer-reviewed) journals, and journal articles to find data for hypothesis testing.

  Here are some databases and search engines to find these types of sources:

  ---

  • Databases for Scientific and Academic Sources

    Scientific Papers

    • ScienceDirect
    • BioOne
    • PubMed
    • PLOS

    Whitepapers

    • Whitepaper Database
    • Tech Republic White Papers
    • AWS White Papers
    • Food-Energy-Water White Papers

    Many companies publish their own whitepapers, so be sure to search for whitepapers on your topic.

    Research Papers

    • ClinicalTrials.gov
    • New York Public Library
    • Agency for Healthcare Research and Quality
    • AVMA Animal Health Studies
    • The University of Iowa Clinical Research
    • Oregon Clinical and Translational Research Institute
    • UC San Diego Health Research

    Surveys

    • Pew Research
    • Deloitte Insights
    • Nielsen Reports
    • JLL Research
    • PwC Research and Insights
    • KPMG Insights
    • American Hospital Association Survey Database
    • Farm Household Survey Database
    • GACP Law Enforcement Surveys

    Academic Databases

    • Google Scholar
    • EBSCO
    • ProQuest
    • JSTOR
    • CORE
    • ScienceOpen
    • DOAJ
    • ERIC
    • arXiv
    • PLOS
    • Paperity

    See more on this page.

  ---

• Use the following three-step approach to test the hypothesis:

  • Step 1: Locate information for both sides of the question using credible sources. While there may be more data for one side or the other, it is critical that you attempt to find data that both proves and disproves the client’s hypothesis.
  • Step 2: Analyze the information found and test the Hypothesis.
    • In many cases, the data will heavily support one side or the other. For these situations, simply present the data and explain that it clearly supports or refutes the hypothesis.
    • If the data is more balanced and there is no clear indication that the client’s hypothesis is true or false, present the data for both sides. Then, using your analytical skills, carefully look at both sides to determine which side outweighs the other, even slightly.
  • Step 3: Based solely on the data, draw a conclusion that either supports or refutes the client’s hypothesis.
    • Be sure to note whether the conclusion is absolute (the data heavily supports one side) or less definite (the data leans toward one side or the other).

Findings

• Organize your findings using bullets and concise headings/subheadings.
• Where possible, corroborate data provided for both sides of the question with multiple sources.

Research Strategy

• If you were able to fully answer all aspects of the project, simply state the sources consulted in your research strategy.
• If you made assumptions or performed any triangulations or calculations, clearly state so in your research strategy.

  • Note: If it makes sense to provide the assumptions, triangulations, or calculations within the findings section, please do so, but briefly summarize the Research Strategy in its own section as well. Here’s an example:

    We used the percentage of millennial smartphone users in Europe and the total number of smartphone users in Europe to calculate the number of millennial smartphone users in Europe. The calculations [or assumptions or triangulations, as applicable] have been provided above [or, in the XYZ section of the report].