\n
Have you ever taken a Buzzfeed-like personality test in which you’re asked questions like:
\n
\n- Which Harry Potter house would you belong to?
\n- What genre of books is your favorite?
\n- What’s your favorite pet?
\n
\n
If you have, then you know that personality assessments can be fun. They can also be invaluable to your business if professionalized.
\n
Personality assessments show your customers that interacting with you is fun and easy. It shows them that you’re interested in learning more about them. And in the process of answering the questions, your customers are able to discover themselves and understand why they make the decisions they do.
\n
The appeal of personality assessments as they relate to business is that it helps you craft unique pieces of content for each personality type or buyer persona. You can offer up this content for download at the end of the assessment or send it via email. This opens up a dialogue between you and the customer. Furthermore, you get contact details like names and email addresses in exchange for the added value you offer through the assessment feedback. Your sales team will thank you for these highly qualified leads.
\n
Knowing your audience’s personalities also helps you create offers and events that are just right for them. For example, an introverted customer might not want to attend a physical event you’re holding, but they may be down to attend a webinar.
\n
\n
How to collect nominal data
\n
Nominal data is usually collected via surveys or questionnaires, often sent our via a cold email marketing campaign. These surveys contain open-ended, multiple-choice, or yes/no questions.
\n
\n- Open-ended questions give the respondents the freedom to describe their opinions or thoughts in their own words without any restrictions. For example, ‘What do you like about living in New York?’ is an open-ended survey question. The answer will likely differ with each respondent, but you’ll be able to get a wide variety of insights that can inform your decision-making.
\n
\n
\n
![\"\"](\"https://proweb.p.blends.be/wp-content/uploads/Scherm_afbeelding-2023-02-23-om-10.16.03-768x428-1.png\")
\n
\n
Multiple-choice questions are set up so that respondents can choose between options. A great example is this seafood illustration: If you had to choose one of these seafood dishes to order in a restaurant, which would you pick?
\n
\n- Curried scallops
\n- Crab cakes
\n- Paella
\n- Shrimp scampi
\n- Fish tacos
\n
\n
This format helps you get specific answers that directly affect your decisions, which facilitates your job in data analytics. Multiple choice questions are easier to pour into an analysis report than open-ended questions.
\n
Yes/No questions are simply questions that require a Yes or No answer. For example, Are you over the age of 18? is a Yes/No question.
\n
How to analyze nominal data
\n
Because it has no numerical value and cannot be put in an order, nominal data can get messy—especially when you’re collecting the data from a large group of people. Fortunately, there are ways to analyze and make sense of nominal data. They include:
\n
\n- Descriptive statistics
\n- Data visualization
\n- Non-parametric statistical analysis
\n
\n
Descriptive statistics
\n
Descriptive statistics are informational coefficients that summarize a dataset found in a study.
\n
A dataset is a compilation of responses and observations from the entire population or a sample of the population. Descriptive statistics describe the features of a dataset and make it easier to understand.
\n
There are three main types of descriptive statistics:
\n
\n- Frequency distribution
\n- Central tendency (mean, median, and mode)
\n- Variability (standard deviation, minimum and maximum variables, variance, kurtosis, and skewness)
\n
\n
For nominal data, you can only use frequency distribution and central tendency (mode only) to analyze nominal data. You can’t use variability because nominal data isn’t quantitative and not subject to mathematical operators.
\n
Frequency distribution
\n
Picture this: You’re planning to open a new seafood restaurant and you’re making your food menu. You have five food choices: curried scallops, crab cakes, paella, shrimp scampi, and fish tacos. So you ask 15 people to tell you which food they’d pick if they had to choose just one.
\n
After gathering their responses, you might notice that the data is disorganized and there’s no way to see how many people picked a specific dish over the others.
\n
To analyze this data, you’ll need to create a frequency distribution table that allows you to see how many responses you get for each food category. A great way to do this in Excel or Sheets is to create a frequency distribution table:
\n
\n
![\"\"](\"https://proweb.p.blends.be/wp-content/uploads/Scherm_afbeelding-2023-02-22-om-16.45.22-768x299-1.png\")
\n
\n
Another example: A marketing consultant wants to learn more about the online presence maturity of her audience. She creates an online presence assessment and waits for the responses to flow in.
\n
The results in this survey dashboard immediately show her how many responses each value has received. She can even choose how the frequency is represented: in absolute or relative numbers, in the form of a bar chart, pie chart, … More on that later.
\n
\n
![\"\"](\"https://proweb.p.blends.be/wp-content/uploads/Scherm_afbeelding-2023-02-22-om-16.46.51-768x387-1.png\")
\n
\n
Central tendency (mode)
\n
Central tendency deals with the average value of a dataset. There are three ways to measure central tendency:
\n
\n- Mode: The value that appears most frequently in the dataset
\n- Median: The middle or center value of the dataset
\n- Mean: The average value of the dataset
\n
\n
With nominal data, you can only use mode to measure central tendency. To identify the mode, look for the category that appears most in your frequency distribution table. In our example dataset, shrimp scampi has the most responses (6 out of 15 people, or 40%). This makes shrimp scampi the mode of the dataset.
\n
Data visualization
\n
Data visualization is the representation of a dataset in the form of a chart, diagram, graph, or map. Visualizing and transforming your data can help you see data trends, patterns, and outliers easily.
\n
Column charts and pie chart designs are the go-to way of visualizing data. You can create them in Excel or Sheets by clicking Insert and selecting Chart from the menu. With a survey app, generating visual representations is automated for you.
\n
\n
![\"\"](\"https://proweb.p.blends.be/wp-content/uploads/Scherm_afbeelding-2023-02-22-om-16.48.40-768x448-1.png\")
\n
\n
![\"\"](\"https://proweb.p.blends.be/wp-content/uploads/Scherm_afbeelding-2023-02-22-om-16.48.49-768x439-1.png\")
\n
\n
Non-parametric statistical analysis
\n
Descriptive statistics and data visualization only summarize nominal data. Statistical tests, on the other hand, allow you to test a hypothesis based on the data you’ve collected.
\n
There are two types of statistical tests:
\n
\n- Parametric tests: These are used for quantifiable data (e.g. interval and ratio data, which we’ll discuss later)
\n- Non-parametric tests: These are used for qualitative data (e.g. nominal and ordinal data)
\n
\n
Since we’re working with nominal data, we’ll use non-parametric statistical tests. There are two types of non-parametric statistical tests:
\n
Chi-square goodness of fit test
\n
If you’re analyzing a dataset with only one nominal variable, the Chi-square goodness of fit test will help you evaluate if your dataset is representative of the whole population.
\n
Let’s use our seafood restaurant example. Imagine that, prior to sending out the survey, you found an official report that showed that most people in your state or country who eat seafood will pick a paella out of all five dishes. But your dataset showed that people preferred shrimp scampi in general.
\n
Since it’s not possible to gather data from everyone in your state, you can use the Chi-square goodness of fit to know how applicable your findings are to the population of seafood-eating people living in your state or country.
\n
Chi-square test of independence
\n
If your dataset has two nominal variables, you can use the Chi-square test of independence (or Chi-square association test) to explore the relationship between both variables.
\n
For instance, say you wanted to find out if there’s a significant correlation between your respondents’ possible allergies and the seafood dish they chose. You’d have to carry out a Chi-square test of independence to compare the frequency of each food choice across the categories of the second nominal variable allergies.
![\"\"](\"https://proweb.p.blends.be/wp-content/uploads/Scherm_afbeelding-2023-02-22-om-16.24.49-768x403-1.png\")
![\"\"](\"https://proweb.p.blends.be/wp-content/uploads/Scherm_afbeelding-2023-02-22-om-16.32.24-768x348-1.png\")
![\"\"](\"https://pointerpro.com/wp-content/uploads/PP-BANNER-GetStartedToday.png\"\n "\\"\\"/")
\n ![\"Understanding](\"https://pointerpro.com/wp-content/uploads/Data-from-survey-with-deviation-scaled.jpg\")
\n ![\"read](\"https://pointerpro.com/wp-content/themes/project-child/assets/img/icons/arrow-right.png\"){kind=icon}
\n ![\"Assessment](\"https://pointerpro.com/wp-content/uploads/Scherm_afbeelding-2023-10-23-om-12.27.36-1024x408-1.png\")
\n ![\"Survey](\"https://pointerpro.com/wp-content/uploads/SURVEYxANYPLACE-92.jpg\")
\n