Giving students a choice in how they represent their data is one way to help students move towards being more self-directed with the use of data; however, they need to understand what graphs best represent the data. In addition, letting students choose how to represent data helps them identify patterns. In this section, we will explore various ways that data can be represented and examine some possible tools that can be used to create the representations.
There are many different ways to represent data. In the airplane data found in Activity 5 of Module 3: Analyzing Data, a line graph was used to compare the wingspan vs. flight distance. Why do you think this graph was the most effective for the representation of that data? There are many data types available to consider as you build your bite-size experience. Think about which data type and corresponding graph might best help represent the data students will collect.
Data that is in discrete sets or categories (nominal data)
This includes data such as male and female and type of animal. Data with separate categories should be represented in bar graphs or pie charts. Individual bars imply there is no connection between the points. No animal can be ½ dog and ½ cat, for example. A study comparing the amount of food consumed by different types of farm animals should be represented on a bar graph.
Data that is in discrete sets or categories that can be ordered numerically (ordinal data)
Sometimes discrete data has an order, such as survey data collected on a Likert scale that has a progression - strongly agree, agree, disagree, strongly disagree. The distance or amount between two categories may not be consistent. In an investigation studying the size of paper airplane wings on the distance the plane flies, the size of the wings would represent ordered data. This type of data might be represented in bar graphs or line graphs.
Data that is measured on a continuous scale (interval-ratio level data)
Time, distance, and mass are examples of this data. This type of data is measured with units and the measurements are proportional against a consistent scale. There are definite points on a scale between measurements. This type of data might also include counts. This type of data can be represented on line graphs, scatterplots with the best fit line or trend lines, histograms, or box plots.
Histograms and box plots show variation in data. This type of representation helps scientists study averages and variations to draw conclusions.