No, descriptive statistics include reporting mean, median, standard deviations,etc. Correlation is bivariate analysis while regression is multivariate analysis.

Descriptive statistics is a vague term and usually applied to describing the properties of a single variable. However there is (edit: no – thank you @bruce) law that says that a correlation cannot be a descriptive statistic. We might make the distinction between descriptive and hypothesis testing statistics, for example.

I would imagine that a table of the intercorrelations of the items of a scale would be descriptive rather than hypothesis testing, for example.

But there's no legal definition, so no worries!

I'd vote with Ronán Michael Conroy on this one. I would say that, e.g., a table of correlations falls under descriptive statistics. But, I don't know when the distinction between descriptive and other-than-descriptive matters.

I agree with Ronán Michael Conroy too, assuming he meant to say:

"However there is no law that says that a correlation cannot be a descriptive statistic."

The "no" is missing in the original. ;-)

Bruce Weaver Thank you – duly corrected!

My sigestion is to call it "Analytical statistics". In that case we should have descriptive, analytical and inferential statistics

@Annu: I'm agree

Oluwaseyi Ayorinde Mohammed , The basic distinction between descriptive statistics and inferential statistics is that descriptive statistics describe something about the observed data, and inferential statistics make an inference about the population that has been sampled.

Calculating a mean is a descriptive statistic. Calculating a confidence interval for the mean is inferential, because it says something about the larger population. Likewise, I would argue, calculating a correlation between two variables is descriptive. But performing a hypothesis test on the correlation is inferential.

Also, practically speaking, I think presenting the correlation among variables is often included as part of the descriptive statistics if that's not the main analysis of the study.

The distinction may be an issue when a researcher tries to set up a hypothesis. Sometimes, researchers set up a hypothesis using correlations between variables. In this context, however, a correlation between variables is not considered a testable hypothesis, although some may disagree. A major reason I say this is that the correlation matrix between variables is required in an APA-style report. One may consider correlation as an inferential statistics but in a research report it is presented to describe the variables in the study.

Yes, correlation is an example of descriptive statistics.

Descriptive statistics involves summarizing and describing the characteristics of a dataset, providing insights into its central tendency, dispersion, and relationships between variables. Correlation specifically focuses on quantifying the strength and direction of the linear relationship between two variables.

Correlation measures the degree to which two variables move together or vary in relation to each other. It provides a numerical value called the correlation coefficient, which ranges from -1 to +1. A correlation coefficient of +1 indicates a perfect positive correlation, meaning that as one variable increases, the other variable also increases in a linear fashion. Conversely, a correlation coefficient of -1 indicates a perfect negative correlation, where one variable increases as the other decreases in a linear fashion. A correlation coefficient of 0 indicates no linear relationship between the variables.

Descriptive statistics, including correlation, help to summarize and describe the data without making inferences or drawing conclusions about the population from which the data is sampled. They provide valuable insights into the patterns and associations present in the data, aiding in the exploration and understanding of relationships between variables.

Also It is possible to answer "No" to the question "Is correlation an example of descriptive statistics?"

However, it is important to note that this answer may vary depending on the context or perspective.

While correlation is often considered a descriptive statistic because it describes the relationship between variables in a dataset, it can also be seen as a tool or method used in inferential statistics. Inferential statistics involves making inferences or drawing conclusions about a population based on a sample of data.

Correlation can be used in both descriptive and inferential contexts. In a descriptive context, correlation summarizes and describes the relationship between variables within the observed data. It helps to understand the patterns and associations present in the dataset.

On the other hand, correlation can also be used in inferential statistics when the goal is to make generalizations about a larger population based on a sample. In this case, correlation is often used to test hypotheses or make predictions about the relationship between variables beyond the observed data.

Therefore, while correlation is commonly associated with descriptive statistics, it can also be used as a tool in inferential statistics, depending on the purpose and context of the analysis.

There is a case to be made for both a "yes" and "no" answer. The case for "yes" is that if you're doing anything correlation in nature (e.g. any regression) you would want to include a simple correlation table in your manuscript. However, the case for "no" is twofold; first, it is not a descriptor of a single variable, obviously, and in many cases a correlation is not useful depending on what you're actually doing with your data. Second, it is a measure you can hypothesis test, i.e. using Fishers z-test, where you can hypothesis test whether two coefficients are statistically different. So I lean on the "no" end of this spectrum, as it cannot be used to describe a single variable and that you can hypothesis test coefficients for differences.

yes, correlation is a simple descriptive statistics

No,

Correlation shows strength of relationship b/n two variables, H.e

Discriptive stastics summarize the characteristics of data set

Unfortunately, we use the same term for descriptive statistics and population parameters.

Any number you compute on known data is a descriptive statistic for that data. Correlation, mean, number of points, largest value--all are descriptive statistics. They are ways of summarizing data, highlighting an aspect you think is important without listing every point.

But correlation (and mean, and standard deviation, and many other descriptive statistics) can also refer to an parameter of an underlying population or process. So we can talk about the historical correlation of high-school GPA and SAT scores among past applicants to a college (descriptive statistic), or we can hypothesize some global relationship among the two variables and estimate its correlation (population parameter)--probably by computing the historical correlation and putting an error band around it.

At the risk of going off on a tangent, I am going to quote some interesting comments by Clyde Schechter in a Statalist discussion (see link at the bottom of this post). That discussion was about the variance (and SD), but I think the same arguments could apply to the correlation between two variables, as it is a function of the covariance and the two SDs. Here is the excerpt with some highlighting (boldface) added by me.

--- start of excerpt ---

Digression: I have always disliked the terms "sample" and "population" standard deviation. They become confusing because, for example, you are in a situation where what you have is clearly a sample, but the appropriate statistic, it seems, is the "population" standard deviation. I think that it would be better to use different names. In my own mind I think of the "sample" standard deviation as the "inferential" standard deviation and the "population" standard deviation as the "descriptive" standard deviation. I think this terminology would better reflect the fact that the "sample" standard deviation is actually an estimate of the standard deviation in the full population inferred from the distribution observed in the sample, where as the "population" standard deviation is a pure measure of the variation observed in the sample, with no generalizability to the larger population.

--- end of excerpt ---

As you can see in posts #6 and #7, a couple of us indicated that we liked Clyde's suggestion.

Source: https://www.statalist.org/forums/forum/general-stata-discussion/general/1381078-calculating-population-standard-deviation, post #2

There is no definitive answer to whether correlation is descriptive statistics, as different sources may have different definitions and classifications of statistics. However, one possible way to approach this question is to say that a correlation coefficient is a descriptive statistic when it summarizes the relationship between two or more variables in a sample, but correlation is not descriptive statistics when it is used to infer or test causal relationships between variables in a population.

Annu Annu Correlation@ measures the degree of closeness of the relationship between 2 or more variables. If the correlation value obtained is not tested statistically (hypothesis testing) then it is descriptive statistics. If the hypothesis testing is carried out on the correlation value, then it is inferential statistics. Likewise for regression. Regression measures the form of the relationship between 2 or more variables

My take is that descriptive statistics are descriptions of the property of samples, e.g. mean, standard deviation, but don't tell one anything about the presumably represented population. They can be used inferentially as when tested for statistical significance or provided with confidence intervals. Thus, correlations are descriptive statistics which can be put into an inferential context.

Descriptive statistics usually refers to the description of the data e.g., measure of central tendency (mean, median, mode) and measure of variability (standard deviation, range, variance). In purely statistical terms, correlation is not included in descriptive statistics. But in research studies (research articles & research thesis), researchers normally consider correlation a part of descriptive statistics. Again, there is no standard practice what to include in descriptive statistics.

Correlation refers to the degree of mutual influence between two or more characteristics within a system, and the mathematical relationship after regression is mathematical model.

Bruce Weaver , saying ""However there is no law that says that a correlation cannot be a descriptive statistic." There could be strange laws out there. e.g., https://www.in.gov/library/files/Pi_Bill.pdf

Yes, correlation is an example of descriptive statistics. It is a statistical indicator of the relationship between variables and describes an association between types of variables.

My argument on this is that descriptive statistics as in measures of central tendency, variability only describe the data while correlation deals with the degree of association between two or more variables.

Ademigbuji Alice can you and others, who made the same suggestion, explain, why a "descriptive" association between two variables (e.g. covariance, correlation) cannot be seen as descriptive statistics? How many mathematical operations are allowed, until it fails to remain "descriptive" and becomes something else? For the covariance you only need the expected values/means, for the correlation additionally the standard deviations. Each have been considered "descriptive" so far, but the combination isn't descriptive anymore? Seems a bit odd to me. What else as a decription of the sample property is a correlation in your view?