5. correlation between interval and any nominal variables (1) (1)

30
SPEARMAN RANK CORRELATION COEFFICIENT (r s )

Upload: olpot

Post on 15-Apr-2017

224 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT

(rs)

Page 2: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

When the variables of interest are measured in an ordinal scale, the

spearman rank correlation coefficient (rs) maybe used instead of the Pearson r. To obtain the Spearman rs , apply the following steps summarized as follows:

Page 3: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

STEP 1. Rank the scores in distribution X giving the lowest a rank of 1 and the highest a rank of n. Repeat the process for the scores in distribution Y.

Page 4: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

STEP 2. Obtain the difference (di) between the two sets of ranks.STEP 3. Square each difference and then take the sum of the squared di

Page 5: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

STEP 4. Complete the formula

Page 6: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

STEP 5. If the proportion of tries in either X or the Y observations is large, use the formula.

*optional

Page 7: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

where:

tx = number of observations in X tied at a given rankty = number of observations in Y tied at a given rank

Page 8: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

STEP 6. To test whether the observed rs value indicates an association between variables , the following maybe applied:For n from 4 to 30, critical values of rs for .05 and .01 level of significance are shown in the table.For n>30, significance of the observed rs under the null hypothesis can be determined using the t-test using the formula:

Page 9: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

The sampling distribution of the test is the student t distribution with n-2 degrees of freedom.

Page 10: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

  Math Rank X Stat Rank

Y di di2

1 18 7 24 4 3 92 17 6 28 6 0 03 14 5 30 7 -2 44 13 4 26 5 -1 15 12 3 22 3 0 06 10 2 18 2 0 07 8 1 15 1 0 0

COMPUTATION FORMAT:

Page 11: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

Page 12: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

This test of significance of the null hypothesis using the computed rs in the example above is:H0: r=0(There is no relationship between math scores and statistics scores of students)Ha: r(There is significant relationship between math scores and statistics scores of students)

Page 13: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SPEARMAN RANK CORRELATION COEFFICIENT (rs)

At α = .05 the tabular t is tα/2.(n-2)=t.025.5=2.571

Decision: The null hypothesis is accepted because the computed t-value is less than the tabular value.Conclusion: The math scores is not significantly correlated to the statistics scores obtained by students.

Page 14: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Correlation Between Interval and Any Nominal Variables

The use of Correlation Rho Formula 

E=∑Niyi

2– Ny2

∑y2 – Ny2 Where: Ni = number of sample

per categoryyi = average obtained

per categoryN = total no. of samplesy = over-all averagey = individual item

y

2

Page 15: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Let us measure the degree of relationship between the civil status and the anual salary of the given samples.

Page 16: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Single 65 83 81 69 73 89 76 60

Married 70 67 90 84 78

Widowed 89 64 78

Page 17: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Solution:N1 = 8 y1 = 596/ 8 = 74.5

N2 = 5 y2 = 389/5= 77.8

N3 = 3 y3 = 231/3= 77

N = 16 y = 1216/16 =76.0

Page 18: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

y2 = (65)2 + (83)2 + (81)2 + ... + (89)2 + (64)2 + (78)2 = 93,792

[8(74.5) 2 + 5(77.8) 2 + 3(77)2] – 16(76) 2E2

= 93,792 – 16(76) 2 E2 =

0.03

Page 19: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

INTERPRETATION: There is a very small positive relationship between the civil status and the annual salary of the given samples

Page 20: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Let us measure the degree of relationship between the subjects and the scores of the given samples.

Page 21: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Science 15 20 9 3 12 16

Math 5 5 14 6

English 23 13 12

Page 22: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Solution:N1 = 6 y1 = 75/ 6 = 12.5

N2 = 4 y2 = 30/4= 7.5

N3 = 3 y3 = 48/3= 16

N = 13 y = 153/13 =11.77

Page 23: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

y2 = (15)2 + (20)2 + (9)2 + ... + (23)2

+ (13)2 + (12)2 = 2239 [6(12.5) 2 + 4(7.5) 2 + 3(16) 2] – 13(11.77) 2E2

= 2239 – 13(11.77) 2 E2 =0.3

Page 24: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

INTERPRETATION: There is a very small positive relationship between the subjects and the scores

Page 25: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Let us measure the degree of relationship between the performance rank obtained by the trainees during the first and second evaluation period.

Page 26: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Student Trainee Rank During 1st Evaluation

Rank During 2nd Evaluation

A 8 7B 2 5C 7 10D 1 4E 4 2F 9 6G 3 1H 6 9I 10 8J 5 3

Page 27: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

SOLUTION:Student Trainee

Rank During 1st Evaluation

Rank During 2nd

Evaluation D D2

A 8 7 1 1B 2 5 -3 9C 7 10 -3 9D 1 4 -3 9E 4 2 2 4F 9 6 3 9G 3 1 2 4H 6 9 -3 9I 10 8 2 4J 5 3 2 4

∑D2= 62

Page 28: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

ρ = 1- 6(62)10(102

-ρ = 1- 37210(102-1)

ρ = 0.62

1)

Page 29: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

Interpretation:

There is a high positive correlation between the

student trainees’ performance rank during the first and

second evaluation period.

Page 30: 5. Correlation Between Interval and Any Nominal Variables (1) (1)

THANK YOU

CREDITS TO: Ian EstradaReported by: Christen Diana Gallinera

Jocelle Mella