Professional Documents
Culture Documents
Spring 2005
Superstitions
Walking under a ladder Opening an umbrella indoors
Empirical Evidence
Consumption of ice cream and drownings are generally positively correlated. Can we reduce the number of drownings if we prohibit ice cream sales in the summer?
Prediction
Variables are assigned independent and dependent roles. Both variables are observed. There is a weak causal implication that the independent predictor variable is the cause and the dependent variable is the effect.
Causal
Variables are assigned independent and dependent roles. The independent variable is manipulated and the dependent variable is observed. Strong causal statements are allowed.
General Overview of Correlational Analysis The purpose is to measure the strength of a linear relationship between 2 variables. A correlation coefficient does not ensure causation (i.e. a change in X causes a change in Y) X is typically the Input, Measured, or Independent variable. Y is typically the Output, Predicted, or Dependent variable. If, as X increases, there is a predictable shift in the values of Y, a correlation exists.
Non-Linear Correlation
As the value of X increases, the value of Y changes in a non-linear manner
No Correlation
As the value of X changes, Y does not change in a predictable manner. Large values of X seem just as likely to be paired with small values of Y as with large values of Y
Interpretation
Depends on what the purpose of the study is but here is a general guideline...
Some of the many Included in SPSS Types of Correlation Coefficients Bivariate Correlation
(there are lots more. these are the ones we will procedure focus on this semester)
Name Pearson r Spearman rho Kendall's Tau Phi Intraclass R X variable Interval/Ratio Ordinal Ordinal Dichotomous Interval/Ratio Test Y variable Interval/Ratio Ordinal Ordinal Dichotomous Interval/Ratio Retest
Both X and Y measured at the Interval/Ratio level Most widely used coefficient in the literature
Computing Pearson r
Hand Calculation
Step #1
Pearson r = 0.73
Step #1
VARX VARY
Output #2:
Correlations VARX VARX Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N 1 . 5 .731 .161 5 VARY .731 .161 5 1 . 5
VARY
Interpretation
r = 0.73 : p = .161
VARX VARY
Correlations VARX VARX Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N 1 . 25 .731** .000 25 VARY .731** .000 25 1 . 25
VARY
Interpretation
r = 0.73 : p = .000
Outliers
100 90 80 70 60 50 40 30 20 10 0 0 10 20 30
Variable Y
r = 0.97
40 50 60 70 80 90 100
Variable X
r = 0.72
Variable Y
Check to see if there has been a data entry error. If so, fix the data. Check to see if these values are plausible. Is this score within the minimum and maximum score possible? If values are impossible, delete the data. Report how many scores were deleted. Examine other variables for these subjects to see if you can find an explanation for these scores being so different from the rest. You might be able to delete them if your reasoning is sound.
r = 0.72
MEDIUM
Variable Y
LOW
HIGH
10
20
30
40
50
60
70
80
90
100
Variable X
100 90 80 70 60 50 40 30 20 10 0 0
LOW
Variable Y
45 40 35 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 Variable X
Variable Y
100 90 80 70 60 50 40 30 20 10 0 0 10 20 30
MEDIUM
Variable Y
80 70
Variable Y
Variable X
60 50 40 30 20 20 30 40 50 60 70 Variable X
100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 Variable X
Variable Y
HIGH
70
80
Variable Y
MEDIUM r=0.86
Variable Y
LOW r=0.55
HIGH r=0.67
10
20
30
40
50
60
70
80
90
100
Variable X
Heres another problem with interpreting Correlation Coefficients that you should watch out for..
140 120
Y variable
100 80 60 40 20 0 0
Men r = -0.21
Women r = +0.22
Men Women
20
40
60
80
100
120
140
X variable
Lower triangular correlation matrix. Values are not repeated. There is also an upper triangular matrix!
Sample Problem
Pincherle and Robinson (1974) note a marked inter-observer variation in blood pressure readings. They found that doctors who read high on systolic tended to read high on diastolic. Table 1 shows the mean systolic and diastolic blood pressure reading by 14 doctors. Research question: What is the strength of the relationship between the two variables?
Pincherle, G. & Robinson, D. (1974). Mean blood pressure and its relation to other factors determined at a routine executive health examination. J. Chronic Dis., 27, 245-260.
Table 1. Mean blood pressure readings, millimeters mercury, by doctor. Doctor ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Systolic 141.8 140.2 131.8 132.5 135.7 141.2 143.9 140.2 140.8 131.7 130.8 135.6 143.6 133.2 Diastolic 89.7 74.4 83.5 77.8 85.8 86.5 89.4 89.3 88.0 82.2 84.6 84.4 86.3 85.9
Research question: What is the strength of the relationship between the two variables?
Option #1: Compute a Pearson r If you do not feel this data meet with assumptions of the Pearson r then Option #2: Convert data to Ranks and then compute a Spearman rho We will be going over how to check the assumptions on Wednesday when we talk about Regression
Table 1. Mean blood pressure pressure readings, readings, millimeters millimetersmercury, by doctor. millimeters mercury, bymercury, doctor. by doctor. Doctor ID 11 2 1 10 2 4 3 10 3 4 14 12 5 12 11 6 5 7 14 8 2 13 9 8 10 6 9 11 9 6 12 8 1 13 7 14 1 7
Systolic Systolic 141.8 130.8 140.2 141.8 140.2 131.7 132.5 140.2 131.8 131.7 131.8 132.5 131.8 132.5 135.7 133.2 135.6 135.7 141.2 135.6 130.8 141.2 135.7 143.9 143.9 135.7 133.2 140.2 140.2 143.6 140.8 140.8 140.2 141.2 131.7 131.7 140.8 140.8 130.8 130.8 141.2 140.2 135.6 135.6 141.8 143.9 143.6 143.6 141.8 133.2 133.2 143.9
(systolic) R (systolic) R(diastolic) Diastolic R Diastolic 89.7 84.6 1 74.4 89.7 8.5 12 14 1 74.4 82.2 2 77.8 74.4 8.5 4 2 1 83.5 3 82.2 83.5 2 3 3 4 77.8 4 83.5 77.8 3 4 4 2 85.8 85.9 5 84.4 85.8 6 7 5 7 86.5 84.4 6 84.6 86.5 11 1 10 6 85.8 89.4 14 7 13 7 89.4 85.8 7 85.9 89.3 8.5 5 12 8 89.3 74.4 8.5 86.3 88.0 13 10 11 9 88.0 89.3 8.5 86.5 82.2 11 2 10 3 82.2 88.0 10 88.0 84.6 10 1 11 6 84.6 86.5 11 89.3 84.4 8.5 6 12 5 84.4 89.7 12 89.4 86.3 14 13 13 9 86.3 13 89.7 85.9 12 5 14 8 85.9 89.4 14
Table 1. Mean blood pressure readings, millimeters mercury, by Table 1. Mean blood pressure readings, millimeters doctor. millimeters mercury, mercury, by by doctor. doctor.
Doctor ID 1 2 3 4 5 6 7 8 9 10 10 11 11 12 12 13 13 14 14 Systolic 141.8 140.2 131.8 132.5 135.7 141.2 143.9 140.2 140.8 131.7 131.7 130.8 130.8 135.6 135.6 143.6 143.6 133.2 133.2 Diastolic Diastolic 89.7 89.7 74.4 74.4 83.5 83.5 77.8 77.8 85.8 85.8 86.5 86.5 89.4 89.4 89.3 89.3 88.0 88.0 82.2 82.2 84.6 84.6 84.4 84.4 86.3 86.3 85.9 85.9 R R(systolic) (systolic) R R(diastolic) (diastolic) 12 14 12 14 8.5 1 8.5 1 3 4 3 4 4 2 4 2 7 7 7 7 11 10 11 10 14 13 14 13 8.5 12 8.5 12 10 11 10 11 2 3 2 3 1 6 1 6 6 5 6 5 13 9 13 9 5 8 5 8
2 d d d ii i -2 4 -2 7.5 56.25 7.5 -1 1 -1 2 4 2 0 0 0 1 1 1 1 1 1 -3.5 12.25 -3.5 -1 1 -1 -1 1 -1 -5 25 -5 1 1 1 4 16 4 -3 9 -3 di = 132.50
Correlations SYSTOLIC DIASTOLI SYSTOLIC Correlation Coefficient 1.000 .708** Sig. (2-tailed) . .005 N 14 14 DIASTOLI Correlation Coefficient .708** 1.000 Sig. (2-tailed) .005 . N 14 14
Spearman's rho
Kendalls Tau (, T, or t)
Named after Sir Maurice G. Kendall
(1907-1983)
Based on the ranks of observations Values range between 1 and +1 Computation is more tedious than rs Defined as the probability of concordance minus the probability of discordance. Typically will yield a different value than rs
To find out more about this statistic, see http://www2.chass.ncsu.edu/garson/pa765/assocordinal.htm
Photo from: http://www.york.ac.uk/depts/maths/histstat/people/sources.htm
SYSTOLIC Pearson Correlation Sig. (2-tailed) N DIASTOLI Pearson Correlation Sig. (2-tailed) N
Spearman's rho
SYSTOLIC Correlation Coefficient Sig. (2-tailed) N DIASTOLI Correlation Coefficient Sig. (2-tailed) N
Kendall's tau_b
SYSTOLIC Correlation Coefficient Sig. (2-tailed) N DIASTOLI Correlation Coefficient Sig. (2-tailed) N
Non-Pearson "family"
Name Kendal's Tau Biserial Tetrachoric Definitions True Dichotomous: A variable that is nominal and has only two levels. Forced Dichtomous: The variable is assumed to have an underlying normal distribution, but is forced to be a dichotomous variable (e.g. Rich/Poor, Happy/Sad, Smart/Not Smart, etc.) Symbol rb rt X Ordinal Forced Dichotomous Y Ordinal Interval/Ratio
From: http://www.oandp.org/jpo/library/1996_03_105.asp
Nonparametric tests should not be substituted for parametric tests when parametric tests are more appropriate. Nonparametric tests should be used when the assumptions of parametric tests cannot be met, when very small numbers of data are used, and when no basis exists for assuming certain types or shapes of distributions (9). Nonparametric tests are used if data can only be classified, counted or ordered-for example, rating staff on performance or comparing results from manual muscle tests. These tests should not be used in determining precision or accuracy of instruments because the tests are lacking in both areas.
From: http://www.unesco.org/webworld/idams/advguide/Chapt4_2.htm
Pearson correlation is unduly influenced by outliers, unequal variances, non-normality, and nonlinearity. An important competitor of the Pearson correlation coefficient is the Spearmans rank correlation coefficient. This latter correlation is calculated by applying the Pearson correlation formula to the ranks of the data rather than to the actual data values themselves. In so doing, many of the distortions that plague the Pearson correlation are reduced considerably.
For more information about the effect of ties on Spearman Rho, see
CONOVER, WJ. Approximations of the Critical Region for Spearman's Rho With and Without Ties Present. Communications in Statistics, Volume B7, No. 3 (1978) (with R. L. Iman), pp. 269-282..