• More Than Staring At Spreadsheets: Making Sense of Data Analytics

    • Share:
    In the information age, consumer data has never before been available in such large
    quantities and in such diversified forms. This means that it has never been easier to generate
    massive spreadsheets that contain equally massive quantities of data, some it useful, most of it
    not so much. Thankfully, new data analytics software has made it easier than ever before to
    quantify data, isolate statistically significant results, and use those results to make business

    I was unaware of the usefulness of data analytics software until I took a course at Saint
    Vincent College under the title of BA 397 – Marketing Research. The class focused on a central
    project, which was a semester-long marketing research project where we sought to gain insight
    into some aspect of our school’s campus. My group chose to pursue assessing the attitudes
    toward a hypothetical new fitness center on campus. Each of us had heard common sentiments
    about the fitness center from word of mouth around campus, but there was no way for us to
    back up student’s claims – we had no data.

    After we defined the problem and developed a purpose for our research, our next order
    of business was to conduct qualitative research. What criteria do students look for in a fitness
    center? How do students evaluate the current facilities? What is the demographic and
    psychographic profile of the current recreation center’s patrons? All of these were questions
    we hoped to address in our focus groups.

    After recruiting volunteers from other classes and holding two separate focus groups for
    student-athletes and non-athletes, we used the students’ responses to construct a survey using
    Qualtrics, a data-collection software. One of the prevailing topics of discussion among both
    focus groups were the windows in the current recreation center. As it currently stands, the
    fitness center has large, glass windows on both the outside and inside walls, allowing passers-
    by to see everyone and everything inside the gym. The windows were a polarizing topic – some
    students said that the windows were a positive addition to the gym because they helped to let
    in additional sunlight and made the room feel larger. Others believed the windows to be a bit
    uncomfortable, producing an effect that made patrons feel like they are in an “observation
    tank.” So, how statistically significant were the student’s claims? To find out, we began
    constructing a data-collection survey in Qualtrics.

    Once the survey was constructed, it was distributed around campus via flyers with QR
    codes printed on them. Our survey had a variety of question formats, the most common of
    which was a five-point Likert scale, which allows survey respondents to rank their approval of a
    given statement from “strongly agree” to “strongly disagree.”

    One of our questions that utilized the Likert scale presented the survey respondents
    with the statement, “The windows in the fitness center make me feel uncomfortable.” Survey
    respondents were also asked to choose their gender in order for us to determine whether there
    were significantly different attitudes between women and men in certain categories. When we
    exported the data into SPSS, a data analytics software, we were able to translate the qualitative
    data from the Qualtrics survey into quantitative data that we could use for our statistical tests.
    The results revealed some interesting new information.

    The above graph shows the combined attitudes of both genders concerning the
    windows. This distribution is bi-modal – not a normal distribution, which meant that about half
    of the respondents strongly disagreed with the statement while the approximated other half of
    respondents tended to strongly agree with the statement. The windows in the fitness center
    were revealed to be a polarizing issue. Thanks to the data analytics software, we were able to
    discover even more information when we filtered the results by gender. Filtering the results by
    “male” produced this graph:

    The male survey respondents were statistically more likely to be in favor of the windows
    in the fitness center. As you may expect, filtering results by “female” produced a near mirror-
    image graph.

    Compared to men, women statistically did not like the windows in the current fitness

    Because of the sentiments we heard during our focus groups, our team had suspected
    that women were generally uncomfortable with the windows while men generally preferred
    them. Now that we had analyze our quantitative data, we had the numbers to back up our
    But why does this matter? Does the intimidation factor affect the amount of time
    people spend in the fitness center? After running a statistical test called a one-way analysis of
    variance (ANOVA), we discovered that it does.

    The uppermost circled number is a number known as the probability value, or more
    simply, the p-value. Generally, in statistics, a null hypothesis is rejected if the p-value is greater
    than 0.05. This low p-value produced by the test indicates that there is a significant correlation
    between intimidation and time spent in the fitness center. This allowed us to come to the
    conclusion indicated by the lower circled number, which is reported in fractions of an hour.
    According to our test, for each additional “unit” of intimidation, patrons spent about 39 less
    minutes in the gym! This is information that could not have been discovered with focus groups
    alone, and it was satisfying to have new data that gave insights into the original claims of our
    focus group.

    Making sense of numbers can be a daunting task, but using data analytics software can
    help aid in decision making by providing valuable insights into the community of our interest.

    Not only can data analytics help to gauge attitudes and feelings toward certain things, but it can
    also reveal new insights that were otherwise unknowable.
    Does deciphering all of this data still seem difficult? Don’t hesitate to stop by the
    chamber and talk with us about how you can use data analytics to make smart decisions and
    gain insights. We might be able to help you – or help you find a member who can!
    Leave a Comment
    * Required field