In this digital age, data is everywhere. We have data on our phones, in our homes, and at work, but how do we know that the data is correct?
The fact is that data can be misleading or even incorrect. In today’s world, with the ability to share and distribute information at the touch of a button, it is easier to spread misinformation.
People can share opinions as facts, and while some of these opinions may be true, they are not necessarily verified information.
How do we verify information then? How do we know that data is accurate? Data can be categorized into three types: discrete, continuous, and categorical. This article will discuss these three categories and how to verify information based on them.
Standard deviation
The third characteristic of data is the standard deviation. This is a measure of how much the data values vary around the average.
The average is also called the midpoint, or median. The median is not always exactly in the middle, but it is always somewhere in between two values.
For example, if we had a list of salaries, the median would be half of the total salaries above $50,000 and half of them being less than $50,000.
The average is calculated by adding up all of the numbers and dividing by how many numbers there are. The standard deviation measures how far each number on average varies from this average value.
This can be visualized with a bell curve (also called a normal distribution). The standard deviation marks out where one quarter of all possible values lie on either side of the average value.
Mean
The mean or average is the most common and most straightforward measure of the variability in a data set. The mean is what most people think of when they hear the word “average.”
Mean is the average value of a data set. Mean can be calculated by adding all of the data values together and dividing by how many data values there are.
For example, suppose a statistics professor has twenty students in her class, and that the grade point average for all of her students is 2.8. The average grade for this class is 2.8, so what does that mean?
It means that on average, each student earned a 2.8 grade. Some students probably got higher grades than that, but on average, the class earned a 2.8. This is an important thing to note- on average, your class earns a certain grade, not every individual student in the class receives that grade.
Deviation
In math, deviation refers to how far a value is from a fixed reference value. For example, the average height of people is 1.75 meters, but any individual person can be taller or shorter than that.
Deviation is also called relative difference, relative distance, or relative variability. In statistics, it is called variance.
Variance measures how far data values are from the data set’s average value. The average value of a data set is also called the midpoint or median.
There are two main types of variance: sample variance and population variance. Sample variance refers to the variance of a sample data set; population variance refers to the variance of an entire population of data values. Both types of variance measure deviation, but sample variation uses a sample average as the midpoint while population variation uses the population’s average as the midpoint.
Histogram
A histogram is a bar chart that groups data into specific intervals and shows the frequency of data within each interval. It is characteristic of a histogram that the intervals are fixed, not variable.
Histograms are useful for determining the average data value, how much data is in one direction versus another, and how widespread the data is.
That is because the bars of a histogram show how many instances of a data value there are per interval. The length of the bar indicates how many instances of that value there are, and the location of the bar indicates which direction most of the values fall in.
The width of the bar can indicate either how spread out the values are, or how many distinct values there are in that interval. Both can be clues as to what type of variable your data represents.
A histogram can be horizontal or vertical, depending on whether the intervals are counted up or down, and whether the bars go left to right or right to left.
Q-Q plot
Next, you should check the data for normality using a Q-Q plot. A Q-Q plot is a way to visualize the distribution of data as a pair of quadrats.
The first quadrat represents the lower quartile (25% of the data) and the median (50% of the data) and the second quadrat represents the upper quartile (75% of the data) and the mean.
If the data is normal, then all of these values should be overlapping. Points that do not overlap indicate some sort of anomaly, usually with one value being too high or too low. An example can be seen in the image above where there is an outlier for the median value.
Using this test alone may not prove whether or not your data is normal, but it can be very informative. It can show you some anomalies in your data and give you an idea of what to look for next.
What is data?
Data is a collection of values or facts that are structured, organized, and shared. Data can be in the form of numbers, words, images, or any combination of these.
Data is typically organized into categories, called data sets. These data sets can be categorized by format (i.e. number, word, image), by source (i.e. government agencies or companies), by use (i.e. for research or for marketing), or any combination of these.
There are many uses for data today, from marketing strategies to research studies and more. Because the use of data is growing rapidly, the demand for quality data is increasing as well. Companies and researchers need access to high-quality data that is organized and valid to use in their work.
But what exactly is valid data? Valid data is raw data that has been verified as true.
Why is data important?
Data is what gives value to the things we do. Data is what tells us whether our efforts are paying off or not. Data is what helps us better understand the world around us and how to adapt to it.
In business, data is a commodity. People pay for access to good data because it helps them make better decisions. In fact, some businesses rely solely on good data to make profits.
There are three characteristics of data that define its importance: quantity, quality, and context. Let’s discuss each of these briefly.
The first characteristic of data is quantity. How much data exists? How many units or values are contained in the data? The more units or values contained in the data, the more valuable it is.
The second characteristic of data is quality. What level of quality exists in the data? Is it accurate, credible, unbiased, trustworthy, and/or reliable? The higher the quality of the data, the more valuable it is.
Characteristics of data
The next level of data is called characteristics. Characteristics are defined as a measure of the amount that the data values vary.
Some examples of characteristics are: average, median, mode, standard deviation, and relative frequency. These all describe different ways to describe a group of data.
Average is a pretty straightforward characteristic. Average is defined as the sum of data values divided by the number of data values.
Median is another characteristic and describes the middle value when all of the data values are ordered from lowest to highest. The average of the two middle values is the median.
Mode is the most frequent data value and can be described as either ordinal or categorical. Mode can be difficult to understand if your dataset contains non-numerical values such as “red” or “apple” which are both modes of “color”.
