16.3 – Data aggregation and correlation

Introduction

Correlations are easy to calculate, but interpretation beyond a strict statistical interpretation, eg, two variables linearly associated, may be complicated — caution is recommended. With respect to interpreting a correlation, caution and temperance is warranted. As previously discussed, “correlation is not causation,” is well known, but identifying when this applies to a particular analysis is not straight-forward. We introduced the problem of two variables sharing a hidden covariation which drives the correlation. In this section we introduce how correlations among grouped (aggregated) data may be quite different from the underlining individual correlations (cf Robertson 1950, Greenland 2001, Portnov et. al 2006).  

Perfect (+1 or -1) correlation?

What if an estimated correlation value between two ratio scale variables turns out to be +1 or -1, the limits of the possible range of the correlation coefficient? Trivially, the reported value of one for the coefficient may be the result of rounding, it was actually 0.97. But, for argument’s sake, let’s assume the coefficient value was 1.0 to as many significant figures as the calculator may report. Another trivial possibility: suspect you’re looking at a case for which the two variables are the same thing, but on different scale. For example, the correlation between a range of temperatures measured in degrees Fahrenheit and then converted to degrees Celsius trivially will be +1.

Less trivially, a correlation of +1 or -1 may reveal that two variables are simply restatements of the same measured variable. Additionally, perfect correlation may reflect construction of a composite variable. Composite variables are examples of indexes, and are constructed from related variables by the researcher in order to better predict multivariate outcomes (see also Latent variables). Refs to add Diamantopoulos and Winklhofer 2001; Bollen and Diamantopoulos 2015

Data aggregation

Data aggregation or grouping refers to processes to group data in a summary form. Considerable public health data is presented this way. For example, CDC reports table after table of data about morbidity and mortality of United States of America population. Data are grouped by age, cities, counties, ethnicities, gender, and states and reports are generated to convey the status of health peoples. Similarly, education statistics, economic statistics, and statistics about crime are commonly crafted from grouped data of what originally was data for individuals.

Correlations between groups may yield spurious conclusions

Researchers interested in testing hypotheses like BMI is correlated with mortality (Flegal et al 2013, Kltasky et al 2017), or health disparities and ethnicity (Portnov et. al 2006), may use grouped data. In 16.2 we introduced the concept of spurious correlation. Correlations between grouped data may also mislead.

Consider the hypothesis that religiosity may deter criminal behavior. This hypothesis has been tested many times dating back to at least the 1940s (reviewed in Salvatore and Rubin 2018). Conclusions about religious beliefs range from negative association with criminal behavior to, in some reports, holding religious beliefs makes one more likely to commit crime. Testing versions of the hypothesis — what causes criminality in some individuals — among a variety of putative causal agents pops up through the history of biology research, arguably beginning with Galton.  I hope you appreciate how challenging this would be to actually resolve — defining criminal behavior itself is laden with all kinds of sociology traps — and for a biologist, reeks of eugenics lore (Horgan 1993).

that all said, let’s proceed to test the religion-criminality hypothesis with aggregated data. The null hypothesis would be no association between crime statistics and numbers of churches. We can also ask about association between crime and non-religious or secular beliefs. I added numbers of Catholic churches and secular humanists groups for cities larger than 100K population by Internet search (FBI for crime statistics, Wikipedia for cities). Figure 1 and Figure 2 report crimes statistics aggregated by cities in the United States and by number of Catholic churches (Fig 1), and by number of secular humanists groups (Fig 2) in the same cities.

Figure 1. Scatterplot crime rates of cities by number of Catholic churches

Figure 2. scatterplot crime rates of cities by number of secular humanist associations.

We’ll just take the numbers on faith (of course, we should think about the bunching around the origin — do we really think Internet search will get all of the secular groups, for example? Or is it really the case that several cities have no secular groups?).  Both correlations were statistically different from zero, crime by churches (P < 0.001) and crime by secular groups (P < 0.001). 

Now, having read Chapter 16.2 I trust you recognize immediately that there’s an important hidden covariate in common. Cities with small populations will have small numbers of crimes reported and smaller numbers of churches compared to large cities. Indeed, the correlation between population and crime for these cities was 0.89 and 0.97, respectively. However, after estimating the partial-correlations, we still have some explaining to do. For crime and churches, the partial correlation was +0.37 (p-value = 0.009); for crime and secular humanist groups, the partial correlation was -0.37 (p = 0.018). These results suggest that persons are more likely to commit crimes in cities with lots of Catholic churches whereas criminal behavior by individuals is less likely where secular humanist groups are numerous.

Before we start pointing fingers, the analysis presented here is a classic ecological fallacy. By grouping the data we lose information about the individuals, and it is the individuals to which the hypothesis applied. Thus, we are at risk of making incorrect conclusions by assuming that the individual is characterized by the group. The hypothesis remains challenging to test (how does one get a valid assessment of an individual’s religiosity? The hypothesis is challenging to test, but studies of individuals tend to find no association or a negative association between criminal behavior and religiosity (Salvatore and Rubin 2018). Crime statistics may underestimate criminal behavior, eg, embezzlement and other “white” crime), but a proper study would look to survey of individuals (Fig 3). 

Figure 3. Illustration of ecological fallacy: positive association at level of groups (boxes, solid blue line), but negative association at level of individuals (black circles, red dashed lines).

Studies that use aggregate data test hypotheses about the groups, not about individuals in the groups. These studies are appropriate for comparing groups, eg, health disparities by ethnicity (Wang Kong et al 2022) or gender (Read and Gorman 2010; Cooper et al 2023), or comparisons among counties (eg, urban vs. rural) for medical resources (Anderson and Zimmerman 2024), or other “social determinates of health” (Crawfolrd 1977; Braveman and Gottlieb 2014), but one cannot conclude that the association is present for members of the group.

Bubble charts

Given we know about the covariation with population size, is there a better way to visualize the crime by number of churches and crime by number of humanist societies? One option is the bubble chart. 

Figure 4. Bubble plot of data used to make Figure 1. Plot by LibreOffice Calc.

Figure 5. Bubble plot of data used to make Figure 2. Plot by ggplot2 package in R.

R code for Figure 5.

# STDHA site by Dr A. Kassambara https://www.sthda.com/english/

myPlot <- ggplot(humanists, aes(x = Humanist, y = Crime.K, size = Pop.M)) + 
geom_point(color="blue", aes(size = Pop.M)) + theme_bw() + ylab("Crime in thousands") + 
xlab("Number of Humanist Societies")

myPlot2 <- myPlot + theme(axis.title=element_text(size=16), 
axis.text.x=element_text(size=12), axis.text.y=element_text(size=12),
panel.grid.major = element_blank(), panel.grid.minor = element_blank())

myPlot2 + scale_x_continuous(breaks=seq(0,30,5)) +
scale_y_continuous(breaks=seq(0,65,10))

Questions

1. Write up three learning outcomes for this page. Hint: Point your favorite generative AI to this page and ask for help.

2. What’s the most likely statistical explanation for why there is an apparent correlation between rates of autism and childhood vaccination rates? See Davidson 2017. 

3. For the following data set, calculate the Pearson product moment correlation and, separately, the Spearman rank correlation. Report the values of the coefficients and provide an interpretation of the results.

v1: 68, 72, 68, 69, 65, 76, 58, 62, 69, 67, 66, 71, 70
v2: 1.727, 1.829, 1.727, 1.753, 1.651, 1.93, 1.473, 1.575, 1.753, 1.702, 1.676, 1.803, 1.778

4. For a small data set on human males ages 24 to 59, the estimated correlation between weight and the calculated BMI index was 0.68. (BMI is an example of a composite variable.) Explain why the correlation between weight and BMI is not +1.

Quiz Chapter 16.3

Data aggregation and correlation

Start Quiz

Question

Confirm

✕

✓

Your answer:

Correct answer:

Next

Please wait..

↺

You got {{SCORE_CORRECT}} out of {{SCORE_TOTAL}}

Your Answers

Chapter 16 contents

• Introduction
• Product moment correlation
• Causation and Partial correlation
• Data aggregation and correlation
• Spearman and other correlations
• Instrument reliability and validity
• Similarity and Distance
• References and suggested readings

16.2 – Causation and Partial correlation

Introduction

Science driven by statistical inference and model building is largely motivated by the the drive to identify pathways of cause and effect linking events and phenomena observed all around us. (We first defined cause and effect in Chapter 2.4) The history of philosophy, from the works of Ancient Greece, China, Middle East and so on is rich in the language of cause and effect. From these traditions we have a number of ways to think of cause and effect, but for us it will be enough to review the logical distinction among three kinds of cause-effect associations:

• Necessary cause
• Sufficient cause
• Contributory cause

Here’s how the logic works. If A is a necessary cause of B, then the mere fact that B is present implies that A must also be present. However, that the presence of A does not imply that B will occur. If A is a sufficient cause of B, then the presence of A necessarily implies the presence of B. However, another cause C may alternatively cause B. Enter the contributory or related cause: A cause may be contributory if the presumed cause A (1) occurs before the effect B, and (2) changing A also changes B. A contributory cause does not need to be necessary nor must it be sufficient; contributory causes play a role in cause and effect.

Thus, following this long tradition of thinking about causality we have the mantra, “Correlation does not imply causation.” The exact phrase was written as early as emphasized by Karl Pearson who invented the statistic back in the late 1800s. This well-worn slogan deserves to be on T-shirts and bumper stickers*, and perhaps to be viewed as the single most important concept you can take from a course in philosophy/statistics. But in practice, we will always be tempted to stray from this guidance. The developments in genome-wide-association studies, or GWAS, are designed to look for correlations, as evidenced by statistical linkage analysis, between variation at one DNA base pair and presence/absence of disease or condition in humans and animal models. These are costly studies to do and in the end, the results are just that, evidence of associations (correlations), not proof of genetic cause and effect. We are less likely to be swayed by a correlation that is weak, but what about correlations that are large, even close to one? Is not the implication of high, statistically significant correlation evidence of causation? No, necessary, but not sufficient.

Spurious correlations

Correlation estimates should be viewed as hypotheses in the scientific sense of the meaning of hypotheses for putative cause-effect pairings. To drive the point home, explore the web site “Spurious Correlations” at https://www.tylervigen.com/spurious-correlations , which allows you to generate X-Y plots and estimate correlations among many different variables. Some of my favorite correlations from “Spurious Correlations” include (Table 1):

Table 1. Spurious correlations, https://www.tylervigen.com/spurious-correlations

These are some pretty strong correlations (cf. effect size discussion, Ch11.4 and 16.1), about as close to +1 as you can get. But really, do you think the amount of cheese that is consumed in the USA has anything to do with the number of PhD degrees awarded in engineering or that apple prices are largely set by the number of lawyers in the USA? Cause and effect implies there must also be some plausible mechanism, not just a strong correlation.

But that does NOT ALSO mean that a high correlation is meaningless. The primary reason a correlation cannot tell about causation is because of the problem (potentially) of an UNMEASURED variable (a confounding variable) being the real driving force (Fig 1).

Figure 1. Unmeasured confounding variables influence association between independent and dependent variables, the characters or traits we are interested in.

Here’s a plot of running times for the best, fastest men and women runners for the 100-meter sprint. Running times over 100 meters of top athletes since the 1920s. The data are collated for you and presented at end of this page (scroll or click here).

Here’s a scatterplot (Fig 2).

Figure 2. Running times over 100 meters of top athletes since the 1920s.

There’s clearly a negative correlation between years and running times. Is the rate of improvement in running times the same for men and women? Is the improvement linear? What, if any, are the possible confounding variables? Height? weight? Biomechanical differences? Society? Training? Genetics? … Performance enhancing drugs…?

If we measure potential confounding factors, we may be able to determine the strength of correlation between two variables that share variation with a third variable.

The partial correlation

There are several ways to work this problem. The partial correlation is a useful way to handle this problem, i.e., where a measured third variable is positively correlated with the two variable you are interested in.

Without formal mathematical proof presented, r_12.3 is the correlation between variables 1 and 2 INDEPENDENT of any covariation with variable 3.

For our running data set, we have the correlation between women’s time for 100m over 9 decades, (r₁₃ = -0.876), between men’s time for 100m over 9 decades (r₂₃ = -0.952), and finally, the correlation we’re interested in, whether men’s and women’s times are correlated (r₁₂ = +0.71). When we use the partial correlation, however, I get r_12.3 = -0.819… much less than 0 and significantly different from zero. In other words, men’s and women’s times are not positively correlated independent of the correlation both share with the passage of time (decades)! The interpretation is that men are getting faster at a rate faster than women.

In conclusion, keep your head about you when you are doing analyses. You may not have the skills or knowledge to handle some problems (partial correlation), but you can think simply — why are two variables correlated? One causes the other to increase (or decrease) OR the two are both correlated with another variable.

Testing the partial correlation

Like our simple correlation, the partial correlation may be tested by a t-test, although modified to account for the number of pairwise correlations (Wetzels and Wagenmakers 2012). The equation for the t test statistic is now

with k equal to the number of pairwise correlations and n – 2 – k degrees of freedom.

Examples

Lead exposure and low birth weigh. The data set is numbers of low birth weight births (< 2,500 g regardless of gestational age) and numbers of children with high levels of lead (10 or more micrograms of lead in a deciliter of blood) measured from their blood. Data used for 42 cities and towns of Rhode Island, United States of America (data at end of this page, scroll or click here to access the data).

A scatterplot of number of children with high lead is shown below (Fig 3).

Figure 3. Scatterplot birth weight by lead exposure.

The product moment correlation was r = 0.961, t = 21.862, df = 40, p < 2.2e-16. So, at first blush looking at the scatterplot and the correlation coefficient, we conclude that there is a significant relationship between lead and low birth weight, right?

However, by the description of the data you should recall that counts were reported, not rates (eg, per 100,000 people). Clearly, population size varies among the cities and towns of Rhode Island. West Greenwich had 5085 people whereas Providence had 173,618. We should suspect that there is also a positive correlation between number of children born with low birth weight and numbers of children with high levels of lead. Indeed there are.

Correlation between Low Birth Weight and Population, r = 0.982

Correlation between High Lead levels and Population, r = 0.891

The question becomes, after removing the covariation with population size is there a linear association between high lead and low birth weight? One option is to calculate the partial correlation. To get partial correlations in Rcmdr, select

Statistics → Summaries → Correlation matrix

then select “partial” and select all three variables (select + Ctrl key; (Windows or macOS) (Fig 4).

Figure 4. Screenshot Rcmdr partial correlation menu

Results are shown below.

partial.cor(leadBirthWeight[,c("HiLead","LowBirthWeight","Population")], tests=TRUE, use="complete")

Partial correlations:
                 HiLead LowBirthWeight Population
HiLead          0.00000        0.99181   -0.97804
LowBirthWeight  0.99181        0.00000    0.99616
Population     -0.97804        0.99616    0.00000

Thus, after removing the covariation we conclude there is indeed a strong correlation between lead and low birth weights.

Partial correlations:
                LowBirthWeight Population
HiLead                 0.99181   -0.97804
LowBirthWeight                    0.99616

Another example

Do Democrats prefer cats? The question I was interested in, Do liberals really prefer cats, was inspired by a Time magazine 18 February 2014 article. I collated data on a separate but related question: Do states with more registered Democrats have more cat owners? The data set was compiled from three sources: 2010 USA Census, a 2011 Gallup poll about religious preferences, and from a data book on survey results of USA pet owners (data at end of this page, scroll or click here to access the data).

This data set also demonstrates use of transformations of the data to improve fit of the data to statistical assumptions (normality, homoscedacity).

The variables, and their definitions, were:

ASDEMS = DEMOCRATS. Democrat advantage: the difference in registered Democrats compared to registered Republicans as a percentage; to improve the distribution qualities the arcsine transform was applied..

ASRELIG = RELIGION. Percent Religious from a Gallup poll who reported that Religion was “Very Important” to them. Also arcsine-transformed to improve normality and homoescedasticity (there you go, throwing $3 words around 🤠 ).

LGCAT = Number of pet cats, log₁₀-transformed, estimated for USA states by survey, except Alaska and Hawaiʻi (not included in the survey by the American Veterinary Association).

LGDOG = Estimated number of pet dogs, log₁₀-transformed for states, except Alaska and Hawaiʻi (not included in the survey by the American Veterinary Association).

LGIPC = Per capita income, log₁₀-transformed.

LGPOP = Population size of each state, log₁₀ transformed.

As always, begin with data exploration. All of the variables were right-skewed, so I applied data transformation functions as appropriate: log₁₀ for the quantitative data and arc sine transform for the frequency variables. Because Democrat Advantage and Percent Religious variables were in percentages, the values were first divided by 100 to make frequencies, then the R function asin() was applied. All analyses were conducted on the transformed data, therefore conclusions apply to the transformed data.

A scatter plot matrix (KMggplo2) plus histograms of the variables along the diagonals shows the results of the transforms and hints at the associations among the variables. A graphic like this one is called a trellis plot; a layout of smaller plots in a grid with the same (preferred) or at least similar axes. Trellis plots (Fig 5) are useful for finding the structure and patterns in complex data. Scanning across a row shows relationships between one variable with all of the others. For example, the first row Y-axis is for the ASDEMS variable; from left to right along the row we have, after the histogram, what look to be weak associations between ASDEMS and ASRELIG, LGCAT, LGDOG, and LGDOG.

Figure 5. Trellis plot, correlations among variables.

A matrix of partial correlations was produced from the Rcmdr correlation call. Thus, to pick just one partial correlation, the association between DEMOCRATS and RELIGION (reported as “very important”) is negative (r = -0.45) and from the second matrix we retrieve the approximate P-value, unadjusted for the multiple comparisons problem, of p = 0.0024. We quickly move past this matrix to the adjusted P-values and confirm that this particular correlation is statistically significant even after correcting for multiple comparisons. Thus, there is a moderately strong negative correlation between those who reported that religion was very important to them and the difference between registered Democrats and Republicans in the 48 states. Because it is a partial correlation, we can conclude that this correlation is independent of all of the other included variables.

And what about our original question: Do Democrats prefer cats over dogs? The partial correlation after adjusting for all of the other correlated variables is small (r = 0.05) and not statistically different from zero (P-value greater than 5%).

Are there any interesting associations involving pet ownership in this data set? See if you can find it (hint: the correlation you are looking for is also in red).

Partial correlations:
         ASDEMS  ASRELIG LGCAT   LGDOG   LGIPC   LGPOP
ASDEMS   0.0000 -0.4460  0.0487  0.0605  0.1231 -0.0044
ASRELIG -0.4460  0.0000 -0.2291 -0.0132 -0.4685  0.2659
LGCAT    0.0487 -0.2291  0.0000  0.2225 -0.1451  0.6348
LGDOG    0.0605 -0.0132  0.2225  0.0000 -0.6299  0.5953
LGIPC    0.1231 -0.4685 -0.1451 -0.6299  0.0000  0.6270
LGPOP   -0.0044  0.2659  0.6348  0.5953  0.6270  0.0000

Raw P-values, Pairwise two-sided p-values:

         ASDEMS ASRELIG  LGCAT   LGDOG   LGIPC   LGPOP
ASDEMS           0.0024  0.7534  0.6965  0.4259  0.9772
ASRELIG  0.0024          0.1347  0.9325  0.0013  0.0810
LGCAT    0.7534  0.1347          0.1465  0.3473  <.0001
LGDOG    0.6965  0.9325  0.1465          <.0001  <.0001
LGIPC    0.4259  0.0013  0.3473  <.0001          <.0001
LGPOP    0.9772  0.0810  <.0001  <.0001  <.0001

Adjusted P-values, Holm’s method (Benjamini and Hochberg 1995)

         ASDEMS ASRELIG  LGCAT   LGDOG   LGIPC   LGPOP
ASDEMS           0.0241  1.0000  1.0000  1.0000  1.0000
ASRELIG  0.0241          1.0000  1.0000  0.0147  0.7293
LGCAT    1.0000  1.0000          1.0000  1.0000  <.0001
LGDOG    1.0000  1.0000  1.0000          <.0001  0.0002
LGIPC    1.0000  0.0147  1.0000  <.0001          <.0001
LGPOP    1.0000  0.7293  <.0001  0.0002  <.0001

A graph (Fig 6) to summarize the partial correlations: green lines indicate positive correlation, red lines show negative correlations; Strength of association indicated by the line thickness, with thicker lines corresponding to greater correlation.

Figure 6. Causal paths among variables.

As you can see, partial correlation analysis is good for a few variables, but as the numbers increase it is difficult to make heads or tails out of the analysis. Better methods for working with these highly correlated data in what we call multivariate data analysis, for example Structural Equation Modeling or Path Analysis.

Questions

1. Write up three learning outcomes for this page. Hint: Point your favorite generative AI to this page and ask for help.
2. Spurious correlations can be challenging to recognize, and, sometimes, they become part of a challenge to medicine to explain away. A classic spurious correlation is the correlation between rates of MMR vaccination and autism prevalence. Here’s a table of numbers for you

3. Make scatterplots of autism prevalence vs
   • Herb supplement revenue
   • Fertility rate
   • MMR vaccination
   • UFC revenue
4. Calculate and test correlations between autism prevalence vs
   • Herb supplement revenue
   • Fertility rate
   • MMR vaccination
   • UFC revenue
5. Interpret the correlations — is there any clear case for autism vs MMR?
6. What additional information is missing from Table 2? Add that missing variable and calculate partial correlations for autism prevalence vs
   • Herb supplement revenue
   • Fertility rate
   • MMR vaccination
   • UFC revenue
7. Do a little research: What are some reasons for increase in autism prevalence? What is the consensus view about MMR vaccine and risk of autism?

Quiz Chapter 16.2

Causation and Partial correlation

Start Quiz

Question

Confirm

✕

✓

Your answer:

Correct answer:

Next

Please wait..

↺

You got {{SCORE_CORRECT}} out of {{SCORE_TOTAL}}

Your Answers

Data sets

Data used in this page, 100 meter running times since 1900.

Data used in this page, birth weight by lead exposure

Data in this page, Do Democrats prefer cats?

Chapter 16 contents

• Introduction
• Product moment correlation
• Causation and Partial correlation
• Data aggregation and correlation
• Spearman and other correlations
• Instrument reliability and validity
• Similarity and Distance
• References and suggested readings