Premium Essay

Correlation Does Not Imply Causation

In: Other Topics

Submitted By 4chani
Words 1350
Pages 6
Correlation does not imply causation

Almost daily we are in (mainly) news media whose owner has a structure similar to some of the following:
One study claims that the more A, the more B
A study says that those who are to have less B.
A study says that since A is, then B is the other way.
In principle, all these headlines indicate that basically what it says A is causing B to happen, or what is the same, that B is a consequence of A. Normally, when one reads the news, just realizing that so there is a correlation between A and B (come on, a relationship between these two events ), but in principle, without any indication that either one of them, even in this case, we cause the other B. (Oakes, 2012)
The study of the correlation between two variables is one of the issues in question in Statistics. To summarize a bit, the question would be something like the following:
- From certain data from each of these variables one estimates if there is any relationship between them. The one most frequently studied is called linear regression (by which we seek if there is no linear relationship between the variables), but there are many more possible types: quadratic, exponential, logarithmic...
- With these data a function (which, for example, is a straight linear regression) that determines us exactly what the relationship between these variables is calculated.
- The actual correlation between them is studied (ie, how strong is the relationship that we calculated based on the initial data) by a correlation coefficient.
This misinterpretation of the correlation is also found, and too often in supposedly serious scientific studies. There are few studies to find some relationship between two variables in the subjects thrown in the pool thus affirming that one causes the other, when in fact in these studies there is no evidence that this is true (simply no correlation).…...

Similar Documents

Free Essay

Tort and Regulations

...Correlation does not imply causation From Wikipedia, the free encyclopedia "Correlation does not imply causation" is a phrase used in science and statistics to emphasize that correlation between two variables does not automatically imply that one causes the other (though correlation is necessary for linear causation, and can indicate possible causes or areas for further investigation... in other words, correlation can be a hint).[1][2] The opposite belief, correlation proves causation, is a logical fallacy by which two events that occur together are claimed to have a cause-and-effect relationship. The fallacy is also known as cum hoc ergo propter hoc (Latin for "with this, therefore because of this") and false cause. By contrast, the fallacy post hoc ergo propter hoc requires that one event occur before the other and so may be considered a type of cum hoc fallacy. In a widely-studied example, numerous epidemiological studies showed that women who were taking combined hormone replacement therapy (HRT) also had a lower-than-average incidence of coronary heart disease (CHD), leading doctors to propose that HRT was protective against CHD. But randomized controlled trials showed that HRT caused a small but statistically significant increase in risk of CHD. Re-analysis of the data from the epidemiological studies showed that women undertaking HRT were more likely to be from higher socio-economic groups (ABC1), with better than average diet and exercise regimes. The use of......

Words: 676 - Pages: 3

Premium Essay

Causation and Correlation

...Causation and Correlation Jennifer PSY/285 Darren Iwamoto July 17, 2013 Causation and Correlation Correlation does not imply causation. According to “statistical Language Correlation and Causation” (Correlation is a statistical measure (expressed as a number) that describes the size and direction of a relationship between two or more variables. A correlation between variables, however, does not automatically mean that the change in one variable is the cause of the change in the values of the other variable.) And (Causation indicates that one event is the result of the occurrence of the other event; i.e. there is a causal relationship between the two events. This is also referred to as cause and effect.) Causation and correlation can be difficult to discern from one another because they are so closely related to one another. Wealthy People are thin. Causation or correlation? The statement “Wealthy people are thin” is a correlation. Not all wealthy people are thin however there may be more thin wealthy people versus non wealthy people due to the fact that wealthy people can afford personal trainers, better food, and healthier lifestyles. People with long hair do better on audio memory tests. Causation or correlation? The statement “People with long hair do better on audio memory tests” is in fact a correlation, it is not a very strong correlation but it is indeed one. Ice cream melts when heated. Causation or correlation? The statement “Ice cream......

Words: 366 - Pages: 2

Premium Essay

Causation and Correlation

...------------------------------------------------- Causation and Correlation ------------------------------------------------- Courtney Clark July 19, 2015 psy/285 Taryn Stevenson July 19, 2015 psy/285 Taryn Stevenson 1. 2. 3. Home >  4. Health & Medicine >  5. Disease 6.  >  7. Causation and Correlation 1. < Back to Health & Medicine Causation and Correlation Correlation does not imply causation * By missweetie * May 11, 2012 * 511 Words * 146 Views There are many similarities between causation and correlation but there are also just as many differences. Causation is when one or more factors contribute to the effect. As said in the PowerPoint review, for example, if you switch a light switch on it causes the light turns on. The one factor of flipping the light switch on causes the effect of the light to turn on. Correlation is when two or more factors contribute to one effect. There is two different types of correlation. One type of correlation is high correlation which is when the factors all match up in a row to cause the effect. Low correlation is when the results of one factor are scattered but a pattern can be recognized. The similarities between causation and correlation are that they both require factors that can point to a result. But remember that correlation is not causation. One factor does not mean it will make the effect happen. The difference between causation and......

Words: 618 - Pages: 3

Free Essay

An Article with Which You Do Not Agree

...Article Analysis An article published by Science Daily outlines a new study which claims that “electronic game use is associated with childhood obesity” (Science Daily, 2004, Par. 1). I disagree with this article because the study seems to break the very basic rule of statistics: correlation does not imply causation. The following will outline why I feel this article makes an unrealistic claim which has not been properly backed up by enough scientific study and experiment. While I agree that the overuse of electronics can lead to a more sedentary lifestyle for children, I do not think the study provided enough information to claim that electronic game use leads to obesity. Instead, it is taking two very common events and then implying that there is causation between the two. The study itself seemed to be performed and presented in a correct and scientific manner, but its results cannot imply a perfect example of causation. For example, the study makes the following claims: “In a study published in the June issue of the Journal of Obesity Research, researchers from The Children’s Hospital of Philadelphia and the University Hospital Zurich present a strong association between playing electronic video games and childhood obesity” (Par. 1). I cannot agree that this claim is valid or should be taken seriously by the reader simply because the study chose two very common events among children and then assumes that they are associated with one another. This would be similar...

Words: 387 - Pages: 2

Premium Essay

Linear Regresion

...of dependence does hours of study have on student’s exam score. 2. Report the r2 linear correlation coefficient and the linear regression equation produced in the Excel spreadsheet. From the excel plot, linear correlation coefficient, r2 or R2 = 0.785 and the linear regression equation is: y = 1.5608x + 55.767 3. What would be the value of Pearson’s r (simply the square root of r2)? Pearson correlation coefficient, r = sqrt(r2) = sqrt(0.785) = 0.886 4. Would Pearson’s r be positive or negative? What does this imply about the relationship between the factors in this study? As evident from the orientation of the scatter diagram (see ex.), r is positive here. r is a measure of the linear correlation (dependence) between two variables x and y, giving a value between +1 and −1 inclusive, where 1 is total positive correlation, 0 is no correlation, and −1 is total negative correlation. 5. What is the implication of any correlation found between the variables in the study you picked? The positive r value implies that there is a trend that one variable (exam scores) goes up as the other one (study hours) goes up. 6. Does this correlation imply a causal relationship? Explain. ‘Correlation does not imply causation’ means that correlation cannot be used to infer a causal relationship between the variables. The Pearson correlation coefficient indicates the strength of a linear relationship between two variables, but its value generally does not completely...

Words: 468 - Pages: 2

Free Essay

Bus 642 Business Research Assignment Two moral choices about our behavior and our relationships with others” (Cooper & Schindler, 2011, p. 32). One of the goals of ethics in research are to make sure that no harm or consequences come from research activities (Cooper & Schindler, 2011). In this scenario, there are several harms and consequences that can occur from the usage of the found information. To begin, there is the problem foresaw by the manager. AutoCorp’s new ethical policy regarding competitor’s information may be pertinent to this information. If an employee in possession of such information is required to return the data or face dismissal, the consequence is not just for the manager (firing) but also for the company (having to replace said employee). This implies a monetary issue for both parties. The manager may rationalize this found information as existing in a gray area of policy, but the question would remain to the consequences of usage. When in question, it would be the wise decision to not risk the consequences and harms that could come to self and others. While the draft found may be only just a draft, there is no true way to tell whether or not the draft has been approved recently, or may be approved by the time it is used. Given that it is also a draft, there may be the possibility that the rival company may decide not to use it at all. That would leave the manager to abandon a legitimate plan for one based on a draft that may or may not be in violation of the new company......

Words: 1934 - Pages: 8

Premium Essay


...0.5 point correctly plotting each ordered pair,for a maximum of 5 points. Part D. Use the scatterplot from Part C to answer these questions. (2 points) What kind of correlation exists between the length and weight of the puppies? Explain. There is a positive correlation exists between the variables, because as the explanatory variable (Length) increases, the response variable (Weigth) increases. Does this correlation imply causation? Explain. It seems to imply causation, since it is logical that the longer a puppy is, the more it would weigh.Students might mention some lurking variables that could affect this conclusion, such as the differences among breeds or the general health of the sampled dogs. Part E. Find the Pearson correlation coefficient (r-value) for the puppy data. Use a graphing calculator or Pearson's formula shown below. Round your answer to the nearest hundredth. (2 points) r = r-value (Pearson correlation coefficient) n = number of values in data set xi = each x-value yi = each y-value = mean of x-values = mean of y-values sx = standard deviation of x-values sy = standard deviation of y-values r-value = 0.93 Part F. Is the correlation between a puppy's length and weight strong? Use your r-value from Part E to explain. (2 points) Yes, the correlation is strongly posively, because the r-value for the data is close to 1. (A strong r-value is 0.8 to 1, or 1, or -0.8 to - 1. These are usually indicative of linear data. ...

Words: 1394 - Pages: 6

Premium Essay

Big Suzy Snack Cake

...results of a regression analysis: * Correlation coefficient (R-squared) – This represents how well the independent variables (X) explain the response variable (Y). * Independent variable coefficient – This is the measured effect the independent variables have on the dependent variable. This is the main output of the regression analysis. * Statistical significance of the coefficient – This is a statistical test that confirms if the coefficient regardless of its value is robust and different from zero. Also referred to as the P-value. Statistical significance of the coefficient The statistical significance of a coefficient tests determines coefficients potential of being zero. The zero potential increases when there is significant variance in the independent variables. A large variance also suggests that the variable used have no effect on the dependent variable. Market Performance As you can see from the previous explanation, it shows that our regression model is highly effective and explains the variation in the number of pies you sold from market to market. The limitation of regression analysis can be described in terms of regression, correlation, and causation. Regression and correlation are related but describe different components of an analysis. While the regression analysis estimates the average value of the dependent variable (Y) on the basis of the fixed values of the independent variable, correlation measures the strength or degree of......

Words: 1333 - Pages: 6

Premium Essay

Demand Estimation and Forecasting

...relationship between variables, correlation analysis examines the strength of the relationship, or goodness of fit. This refers to how closely the points fit the line, taking into consideration the units of measurement. Some idea of this can be obtained from a visual inspection of the graph, but it is better to use a quantitative measure. Correlation. More specifically the correlation coefficient (r) measures the degree of linear association between variables. It should be noted that correlation says nothing about causation. The causation between the variables could be reversed in direction, or it could act in both directions in a circular manner. For example, high sales could lead to economies of scale in production, enabling firms to reduce their price. An alternative explanation of correlation between variables is that there may be no causation at all between two variables; they may both be influenced by a third variable. It should also be stressed that correlation only applies directly to linear relationships, meaning that weak correlation does not necessarily imply a weak relationship; there might be a strong non-linear relationship. Thus drawing a graph of the data is important, since this can give an insight into this possibility. The formula for calculating the correlation coefficient can be expressed in a number of ways, but probably the most common is: The coefficient of determination The problem with the correlation coefficient is that it does not have a......

Words: 1676 - Pages: 7

Premium Essay

Chapter 1 Econ Notes

...Chapter 1 1. Ideas for beyond the final exam a. How much does it really cost? i. Opportunity Costs 1. Value of the next best alternative that must be given up because of that decision. 2. Economists say that the true cost of such decisions are not the number of dollars spend on the computer, new equipment, or military, but rather the VALUE OF WHAT MUST BE GIVEN UP TO ACQUIRE THEM. 3. EXAMPLE- cost of college is more expensive add the money you could be making with a job at home, don’t add housing since you pay that either way (tuition + possible income if you weren’t working) b. Attempts to Repeal the Laws of Supply and Demand- The Market Strikes Back ii. Short Supply-Prices Rise iii. Large Quantity-Prices Fall 4. Ceiling on Price or Price Floors a. Such attempts to repeal the laws of supply and demand usually backfire and cause more problems c. Comparative Advantage iv. Even if one of them is more efficient at everything, both countries can gain by producing the things they do best COMPARATIVELY v. Even if you are better at something than someone else, the time you could be spending on what you specialize in is losing money. Do what you do best and hire someone to do the rest. d. Trade is a Win-Win Situation vi. Fundamental Idea- Both parties must expect to gain something! vii. Some things like ticket......

Words: 992 - Pages: 4

Premium Essay


...might suggest that the treatment or finding does not make enough of a difference to justify its use or to be practical. Yes, it is possible for a study to have statistical significance but not a practical significance. If the source of the data can benefit from the results of the study, it is possible that an element of bias is introduced so that the results are favorable to the source. A voluntary response sample is a sample in which the subjects themselves decide whether to be included in the study. A voluntary response sample is generally not suitable for a statistical study because the sample may have a bias resulting from participation by those with a special interest in the topic being studied. Even if we conduct a study and find that there is a correlation, or association, between two variables, we cannot conclude that one of the variables is the cause of the other. There does appear to be a potential to create a bias. There does not appear to be a potential to create a bias. There does not appear to be a potential to create a bias. There does appear a potential to create a bias. The sample is a voluntary response sample and is therefore flawed. 2. 3. 4. 5. 6. 7. 8. 9. 10. The sample is a voluntary response sample and is therefore flawed. 11. The sampling method appears to be sound. 12. The sampling method appears to be sound. 13. Because there is a 30% chance of getting such results with a diet that has no effect, it does not appear to have statistical......

Words: 3143 - Pages: 13

Premium Essay

Final Exam Psy 325: Statistics for the Behavioral and Social Sciences

...conduct a follow up study to this one. Explain your answer. • Describe the difference between practical and statistical significance. Essay 2 A researcher has investigated the relationship between IQ and grade point average (GPA) and found the correlation to be .75. For this essay, critique the results and interpretation of a correlational study. • Evaluate the correlational result and identify the strength of the correlation. • Examine the assumptions and limitations of the possible connection between the researcher’s chosen variables. • Identify and describe other statistical tests that could be used to study this relationship. Your essay response must address the following questions: • How strong is this correlation? o Is this a positive or negative correlation? o What does this correlation mean? • Does this correlation imply that individuals with high Intelligence Quotients (IQ) have high Grade Point Averages (GPA)? • Does this correlation provide evidence that high IQ causes GPA to go higher? o What other variables might be influencing this relationship? • What is the connection between correlation and causation? • What are some of the factors that affect the size of this correlation? • Is correlation a good test for predicting GPA? o If not, what statistical tests should a researcher use, and why? Essay 3 A researcher has recorded the reaction times of 20 individuals on a memory assessment. The following table indicates the individual......

Words: 807 - Pages: 4

Premium Essay


...details to focus on the most important elements of a problem * The proper degree of abstraction depends on the objective of the analysis * Theory * Deliberate simplification of relationships whose purpose is to explain how those relationships work * Concern for policy that makes economic theory so necessary and important * If you have no idea how something works, how can you fix it? “Getting frustrated and hitting against the table is not the most logical approach.” –Dr. You need to know how things work, or else you can’t fix them. * If you don’t know how the economy works (theory), you cannot fix your company’s demise. * Correlation and Causality * Two variables are correlated if they tend to move together. * Correlation does not imply causation * Economic Model * A simplified, small-scale version of some aspect of the economy * Often expressed in equations, by graphs, or in words * Economics (repetition means something is important) * The study of how individuals and societies allocate their limited resources to try to satisfy their unlimited wants (repeated from above) * Social science which studies human behavior (added) * Basic Choices * What should we produce? * Basket of goods and services * How should we produce it? * The mix of resources * Who will consume it? * The distribution issue *......

Words: 372 - Pages: 2

Free Essay

Bmgt230 Review Sheet

...(color of eyes) Time Series Data- Values over time Cross Sectional Data -specific point in time Interval Data – no meaningful 0, cant divide or multiply but the difference b/w 2 values is meaningful. Ratio Data – opposite of interval, can compare to 0 (height). Selection Bias – sampling scheme problem, systematic tendency to exclude one kind of individual (call at 2 pm everyday). Non Response- subjects don’t answer Response Bias- Subjects lie, interviewer effect Under coverage Bias- portion of population is not sampled. Sampling Error- sample to Sample differences are reflected Sample- part of the population of interest. We examine & do have data Sample Size- fraction of pop u have sampled doesn’t matter; sample size does. Parameter- a number describing a characteristic of population (average student debt) something of interest in the population (unknown). Statistic- a number describing a characteristic of a sample (known) comes from samples Sampling Frame- list of population (phone book, register voter list) Effective population, SRS picks equally from whole frame Statistical Sampling- Simple Random- each individual in pop has equal chance of Being selected Stratified Random- divide pop into sub groups (strata) according to common characteristic, do SRS from each sub group Cluster- groups are a mix of diff people, take a SRS of clusters and look at all individuals in chosen cluster. Data – values along w/ context Variables – the......

Words: 912 - Pages: 4

Premium Essay

Social Psychology and Multicultural Psychology Report

...statistically, the distinction relates to method, rather than the topic studied. There are three main types of psychological research: 1. Correlational research In statistics, dependence is any statistical relationship between two random variables or two sets of data. Correlation refers to any of a broad class of statistical relationships involving dependence. Familiar examples of dependent phenomena include the correlation between the physical statures of parents and their offspring, and the correlation between the demand for a product and its price. Correlations are useful because they can indicate a predictive relationship that can be exploited in practice. For example, an electrical utility may produce less power on a mild day based on the correlation between electricity demand and weather. In this example there is a causal relationship, because extreme weather causes people to use more electricity for heating or cooling; however, statistical dependence is not sufficient to demonstrate the presence of such a causal relationship (i.e., correlation does not imply causation). 2. Descriptive research Descriptive research is used to describe characteristics of a population or phenomenon being studied. It does not answer questions about how/when/why the characteristics occurred. Rather it addresses the "what" question (What are the characteristics of the population or situation being studied?) The characteristics used to describe the situation or population are......

Words: 783 - Pages: 4