Data interpretation refers to the process of subjecting data to predefined processes such as the organization of tables, charts, or Graphs so that logical and statistical conclusions can be derived. This part of Statistics answers a common question among researchers: what exactly are we supposed to present?
It is not ideal for researchers to present numerical values of data collected from instruments or surveys. Data need to be organized to tell the story of what you want to emphasize in your research. This should focus on the problem you want to solve - also known as 'the statement of the problem'. It is the primary function of the research.
Statistical tools are used in the process, helping you to transform data into useful information that can help you to arrive at important conclusions. This process is called data analysis. It is after this process that data can be fully interpreted.
Statistical methods
Statistical methods allow you to work on your data. Imagine you have the exam scores for 100 students, and you want to interpret this data. Scanning through the scores by eye alone might be quite tough! Here are two methods that would simplify this task.
Measures of central tendency
Central tendency values are used to describe some key characteristics of the whole data set, producing a single value that is typical of the whole set. For example, the mode will give you the value that occurs the most often.
The mean is the most commonly reported measure of central tendency and it is the mathematical average. To calculate your mean, you add up all of your values available and divide that by the Number of values you added. The mean is represented by μ, and its formula is, where n is the Number of data items in the sample and is the sum of all data values.
Default deviation
Another statistical measure that is commonly used is variability, also known as spread. The range is the simplest form of variability. Let's take the exam score dataset again - the range is the span between the lowest and highest numerical values.
Another common measure is variance; which is the squared average deviation from the mean. This number indicates how much the individual values deviate from the mean. What you will see reported more often is the Standard Deviation. This is modelled as the square root of the variance. Standard Deviation expresses how much individual class scores differ from the mean value for the group. Mathematically, it can be modelled into an equation:
Single variable data
Single Variable Data involves examining one particular variable relevant to a dataset. Single data analysis is common in descriptive forms of analysis and uses Histograms, frequency Distributions, and Box Plots among other methods. This is mostly used in the first step of investigating data. Let's take a look at a box plot.
boxplot
A box plot displays a five-number summary of a dataset. They are the minimum, first quartile, median, third quartile, and maximum. Quartiles tell us about the spread of data by breaking the data set into quarters. The lower quartile, Q1 represents 25%, the middle quartile equals 50% and the upper quartile represents 75%.
The ages of 10 students in grade 12 were collected and they are as follows:
15, 21, 19, 19, 17, 16, 17, 18, 19, 18.
Let's first arrange these in ascending order.
15, 16, 17, 17, 18, 18, 19, 19, 19, 21.
We can now find the median, which is the middle number. And since we have an even number, we have two of them. Finding the average is standard practice; however, with ours, we have the same number.
We will find the quartiles now. The first is the median to the left of the overall median.
That will mean we are finding the median for 15, 16, 17, 17, 18.
This equals 17.
The third quartile will be the median to the right of the median.
18, 19, 19, 19, 21
Now we will document the minimum number which is 15.
And also document the maximum which is 21.
Figure 1. Box plot
The image above is the box plot representing the data of the ages of the students in grade 12.
We will take another example with an odd number of data points.
The table below is data of basketball players' points scored per game over a seven-game span. Visualise this on a box and whisker plot.
| Game | Points |
| 1 | 10 |
| 2 | 17 |
| 3 | 5 |
| 4 | 32 |
| 5 | 16 |
| 6 | 18 |
| 7 | 20 |
Step 1.
Rearrange the values in the data set from lowest to highest.
5, 10, 16, 17, 18, 20, 32.
Step 2.
Now identify the highest and lowest values in the data set
Highest value: 32
Lowest value: 5
Step 3.
We can now identify the midpoint value (median) of the data set.
Median = 17
Step 4.
We will now find the upper and lower quartiles.
The lower quartile is the median for the first half of the data set.
That will mean that we are finding the median for 5, 10, 16
Lower quartile = 10
The upper quartile is the median for the second half of the data set.
That will also mean that we are finding the median for 18, 20, 32
Upper quartile = 20
Step 5.
Now that we have all our necessary values, we will construct our box and whisker plot.
Highest value = 32
Lowest value = 5
Median = 17
Upper quartile = 20
Lower quartile = 10
We will first draw a Number Line that fits the data, and plot all the necessary values we found.
Figure 2. Plotting necessary values on a box
Construct a rectangle that encloses the median of the entire data set that its vertical lines pass through the upper and lower quartiles. Now construct a vertical line through the median that hits both ends of the rectangle.
Figure 3. Box and plot
There, we have our box and whisker plot for the basketball games.
Bivariate data
In contrast to Single Variable Data, Bivariate Data consist of two variables for each individual. For example, in large studies in the health sector, it is common to collect variables such as height, age, blood pressure, etc. in each individual. Let's look at an example in a two-way frequency table.
These are the number of males and females who had each grade on a math project in school.
degrees | | Female | Total |
A | 9 | | 21 |
B. | 18 | | 32 |
C | | 11 | |
D | 2 | 3 | 5 |
E | 1 | 2 | 3 |
Total | 38 | 42 | 80 |
We can see there are 9 males and 12 females who got an A, 18 males and 14 females who got a B, and so on.
Now we can answer a couple of questions.
How many students in total had an A?
How many males were surveyed?
Answer: 38 males.
How many males earned a grade of A?
Answer: 9.
Below is a graph representation of two variables, the sales of ice cream in a given shop against the temperature of the day. This demonstrates how much ice cream is purchased at every given temperature.
Bivariate Data; Ice cream sales versus temperature of day
Probability
Probability is the measure of how likely an event is to happen. Probabilities can be placed on a Number Line between 0 and 1, as shown below.
So if the Probability of an event is zero, then it is impossible for the event to occur. Whilst if it is 1, then it is certain. Then we have variant degrees in between the two values, and 0.5 would mean there is an even chance of the event happening.
Probability on a number line
Probabilities are written down using the following Notation :
If event A has a between happening and not happening, then the probability of event A not happening = 1 - P (A ')
For example, if the P (A) = 0.8
P(A') = 0.2.
They should both add up to 1.
These are the basic concepts you would be using throughout probability at this level. You can be reintroduced to Venn diagrams, tree diagrams, etc. as well!
Data Interpretation - Key takeaways
- Data interpretation refers to the process of subjecting collected data to predefined processes so logical and statistical conclusions can be derived.
- Presentation refers to the representation of data in Graphs, plots, frequency tables, etc.
- The measure of central tendency produces a single value that is typical of the whole set. The basic values are mean, mode and median.
- Single variable data involves examining one particular variable relevant in a dataset.
- In contrast to single variable data, bivariate data consist of two variables for each individual.
- Probability is the measure of how likely an event is to happen.
Explanations 
Exams 
Magazine 
