Statistics is the branch of mathematics that studies variability, as well as the process that generates it, following the laws of probability, today we will talk specifically about descriptive statistics in psychology.
Statistics are needed both to conduct research and to understand how this research is being conducted today, well beyond the conclusions of any study.
- Thus.
- The knowledge of this branch will allow us to assess the quality of a study and.
- Therefore.
- The degree of reliability of its conclusions.
Descriptive statistics are the part of the statistic that collects, presents, and characterizes a dataset, that is, descriptive statistics try to figure out what happened, unlike inferential statistics, that attempts to predict what will happen in the future under a set of conditions. .
For example, these conditions are usually specified by variables such as age, climate, or degree of anxiety, so descriptive statistics in psychology aim to summarize, usefully for the researcher and reader, what happened in a given study.
As mentioned above, are variables one of the central axes of descriptive statistics?And also not descriptive. A variable encompasses a set of values, and depending on what these values look like, we can talk about:
Variables can also be classified into
Therefore, the data (numbers or measures collected from observation) can be of two types:
The measure involves linking abstract concepts to empirical indicators. The result of this process is called a measure.
There are four possible measurement scales, which are used to assist in the classification of variables, in this sense, reliability and validity properties are very important in descriptive statistics, because they speak of the quality of the measurement.
After all, what is the purpose of data that was poorly collected at its source?
This scale assigns numbers to categories that do not require order (one category cannot be said to be larger than another), and these categories are mutually exclusive.
An example could be sex or color. Thus, the chosen option would be the exclusion of others, this scale is used for qualitative variables.
Here the categories are set with two or more levels that involve an order between them, as on the previous scale, these are also mutually exclusive categories, but now we can place the values of the variables in an order.
For example, the scale can be seen in questionnaire responses
These answer options can be encoded with numbers ranging from one to five, suggesting a default order; however, we cannot know, at least without the help of advanced statistical procedures to estimate it, the distance between two categories.
Thus, we can say that the object of research has more or less of something, but in a simple way, we cannot talk about how much greater that something is (intelligence, memory, anxiety, etc. ).
This scale is also used for qualitative variables
On this scale the distance between the values is quantified, the distance measurement also shows the characteristics of the previous two measures, so it sets the distance between one measure and another.
The range scale applies to continuous variables. However, it is not possible to have an absolute zero on this scale. A clear example of this type of measurement is a thermometer. When you dial zero degrees, it means no temperature.
This scale applies to quantitative variables
Finally, this scale combines the characteristics of the above, determines the exact distance between the ranges of a category, and has an absolute zero point at which there are no characteristics or attributes to measure.
For example, the number of children: zero children means there are no children. This scale applies to quantitative variables.
A frequency distribution is a list of possible values (or ranges) that a variable can take, as well as the number of observations for each value.
This frequency distribution is usually represented by tables, so you must include all possible values for a variable, and you must also indicate the total number of observations (n) that have been made.
When we have a lot of data categories and some of them have very low frequencies, it can be interesting to group them into intervals.
Finally, statistical indicators are used to describe a dataset by a number, so this number summarizes a characteristic of the distribution of the analyzed data, some of these indicators are:
Thus, using these concepts, descriptive statistics organizes, calculates and analyzes the representations of the data to provide the researcher and, by extension, the scientific community with a complete mapping of what happened in his study.