Descriptive Statistics in SPSS or Stata: Understanding their Importance in Irish Research

Students and researchers often seek help or grinds from us with descriptive statistics using various statistical softwares like SPSS, Stata, Excel, R etc. This article first outlines the basics of descriptive statistics using SPSS or Stata. After this, it emphasises the increasing importance of descriptive statistics using SPSS or Stata in Irish research.

Statistical Package for the Social Sciences, referred as SPSS, is often used by students and researchers of various disciplines for data analysis. SPSS is one of the easiest and smartest softwares to perform some quick and robust descriptive statistics. The same goes for Stata in terms of ease of work and robust descriptive statistics analysis. When students and researchers seek help or grinds in descriptive statistics using SPSS or Stata from us, the first question we often ask is, “describe your data to us?”.  As simple as this question may sound, most students and researchers often don’t even get close to providing an appropriate description of their data.

Understanding Descriptive Statistics in SPSS or Stata from their Importance

To the above point on describing data, perhaps the most efficient way to express your data is by performing appropriate descriptive statistics in SPSS or Stata. In other words, descriptive statistics in SPSS or Stata are nothing but specific data values that describe your data.  This data description is needed for both scientific as well as general audience. For example, a dataset in SPSS or Stata format is collected on testing athletes’ performance after some counseling. In such a scenario, one does need to describe in their data about athletes’ age, educational qualification, years of training, minutes of exercise per week etc. If a student or researcher needs to define these variables, descriptive statistics in SPSS in Stata are necessary. This is because the researcher would want to know the average age of sampled athletes or what percentage of athletes have tertiary education, and so on.

Descriptive Statistics in SPSS or Stata: Various Types

Measures of Central Tendency: Mean, Median & Mode

Descriptive statistics in SPSS, Stata, or any other statistical software basically comprise measures of central tendency and variability. Both concepts are easy to understand from a statistical perspective. As with the name, measures of central tendency emphasise “central” or “middle/average” values in the data. For example, in our model of athletes’ age, we want to know the “central” or “average/middle” age of athletes. The common way to do this is to take the mean, median or mode of the data.  If we have collected data from 5 athletes aged 20, 25, 32,32 and 45, the way to calculate the mean is:

Hence, the mean age of our sampled athletes is 30.8 years. Irrespective of statistical software, either SPSS or Stata, we will get the same mean. Median, on the other hand, implies a middle value in the data set. If we arrange our data in ascending order and then pick the middle value, it will be our median. In our data, the median is 32. In contrast, mode implies the most common data point. In our case of athletes age, we can see that there are two athletes aged 32. Therefore, the mode in our case is 32. As with mean, both median and mode also do not change irrespective of using SPSS or Stata.

Descriptive Statistics (Central Tendency) in Stata

Let us open a new Stata data file (.dta) to do these descriptive statistics in Stata. You can type in this data or import it using an Excel file. If we have to do these descriptive statistics in Stata, the easiest way is to use Stata command “sum”, which is the short form of Stata command “Summarise”. The variable age should follow this. Below output shows these results:

As shown in Stata output, the mean we get is 30.8, the same as we estimated manually. To calculate other descriptive statistics like median and mode, we need to use more commands in Stata. So now in Stata, we can follow our “sum” or “summarise” command by “detail” as below:

The above descriptive statistics output in Stata shows detailed descriptive statistics. To recognize the median from this, use the value of 50% percentiles, which shows the median to be 32. An important thing to remember while seeing so many descriptive statistics values in Stata is to concentrate on median only. This is because we will cover other descriptive statistics estimated in Stata in a separate blog.

To estimate descriptive statistics like mode, we need to use another command in Stata as both “sum” and sum followed by “detail” command in Stata do not provide it. To do those descriptive statistics in Stata, use the command “tab” in Stata, which is the short form of command “tabulate”. The variable name should follow this to get the descriptive statistics right. Below output of Stata shows these results:

As seen in the above descriptive statistics output in Stata, the frequency of age “32” is 2, whereas the frequency of other ages is just 1. This implies that 32 is the most common age, and therefore, validated by Stata output, the mode is 32.

Descriptive Statistics (Central Tendency) in SPSS

There are actually numerous approaches to perform descriptive statsitics in SPSS. For measures of central tendency, i.e. mean, median and mode, we can go to Analyse-Descriptive Statistics-Explore. By putting our variable in the dialogue box, we can get the descriptive statistics output desired. See output below:

In SPSS, we have now got mean age as well as median age. The important thing to note here is that the descriptive statistics here are the same as we estimated in Stata. Hence, it is not the statistical software but the understanding of statistics in SPSS and Stata that matter.

As with the case of Stata, descriptive statistics in SPSS also don’t provide us with the value of mode. To estimate that in SPSS, let us go to Analyse- Descriptive Statistics-Frequencies. After putting our variable in Frequency dialogue, we can get the wished output in SPSS. This is shown below:

As shown in the SPSS output, there are two athletes aged 32, and therefore, this is the mode of our data. The same descriptive statistic of 32 we got while estimating in Stata.

Descriptive statistics cover measures of central tendency and measures of variability, whose description and estimation we will cover in both SPSS and Stata statistical softwares, in a different blog.

Descriptive Statistics in SPSS or Stata in the context of Irish Research

Our company, Data Analysis Ireland, frequently receives requests from Irish students and researchers to provide them with help or grinds in Stata or SPSS for descriptive statistics. Our highly qualified experts have helped hundreds of researchers in descriptive statistics and inferential statistics, and for help or grinds in these, our experts can be contacted at support@dataanalysis.ie or WhatsApp number +353 89 278 9288.

However, as our experts have felt, descriptive statistics, especially the measures of central tendency, have garnered massive interest in Irish research, especially in the last ten years. The primary reason we believe descriptive statistics in SPSS or Stata have gained so much interest is the rise of quantitative research in Irish universities. Let’s look at all top-ranking Irish universities like TCD, NUIG, UCC, University of Limerick, DIT etc. Virtually every university has a significant percentage of disciplines focused on quantitative research. With a strong focus on quantitative research, the need for statistical softwares like SPSS or Stata has also increased, and indeed the need for help or grinds in SPSS or Stata, which we provide. Overall, our experts feel descriptive statistics in SPSS or Stata has a strong future in Ireland for decades or perhaps centuries to come.