 Measures the strength of correlation
between a dichotomous (1,0) variable
and a continuous variable (R).
Dichotomous Variable Continuous
Variable
Expected Results
Sex (male (1), Female
(0))
IQ score Inferences about the
relationship between sex
and IQ
Education ( private (1),
public ((0)
Math score Inference about the type
of schooling and
performance on Math
Marital Status (Married
(1), Single (0))
Cholesterol Level Inferences about marital
status and health
Location of residence
(Urban(1), rural (0))
Level of literacy Inferences about location
of residence and literacy
College degree (
graduated (1), no
college degree (0))
Income Inferences about level of
education and income
 Measures the strength of correlation
between a dichotomous variable (1,0)
and an ordinal variable (rank).
Dichotomous Variable Ordinal Variable Expected results
Sex (male 1, female 0) TV networks
preferences
Inferences about sex
and preferred TV
networks
Education ( private 1,
public 0)
TV program
preferences
Inferences about type
of schooling and
preferred TV programs
Marital Status ( married
1, single 0)
Senatorial Candidate
preferences
Inferences about
marital status and
preferred senatorial
candidate
Location of residence
(Urban 1, rural 0)
Bb Pilipinas candidates
preferences
Inferences about
location of residence
and preferred BB.
Pilpinas candidates
 Measures the strength of correlation
between a dichotomous variable (1,0)
and another dichotomous variable (1,0).
Dichotomous Variable Dichotomous
Variable
Expected Results
Sex (male (1), Female
(0))
College Education
(Graduate 1, Non-
graduate 0)
Inferences about sex and
educational attainment
Education ( private (1),
public ((0)
Sports involvement
(involved 1, not
involved 0)
Inference about the type
of schooling and sports
involvement
Marital Status (Married
(1), Single (0))
Financial status
(Poor 1, not poor)
Inferences about marital
status and affluence
Location of residence
(Urban(1), rural (0))
Financial status
(Poor 1, not poor
0)
Inferences about location
of residence and
affluence
College degree (
graduated (1), no
college degree (0))
Employment
(Employed 1, not
employed 0)
Inferences about level of
education and
employment
Name X Y statistic
Point Biserial
Coefficient
Dichotomous
variable
Continuous
Variable
Rank Biserial
Coefficient
Dichotomous
variable
Ordinal
Variable
(Ranking)
Phi
coefficient
Dichotomous
variable
Dichotomous
variable
pbr
rbr
Let X be a dichotomous variable and Y the
continuous Variable
 1
01
1
01



 nn
nn
s
MM
r
n
pb
 
 


n
i
in YY
n
s
1
2
1
1
1
Sample standard
deviation
M1 – the mean value of
observation in the continuous
variable Y for group 1
M0 – the mean value of
observation in the continuous
variable Y for group 2 (0)
n1 – size of sample for group 1
n0 – size of sample for group 2 (0)
n – total sample size

Other correlation coefficients

  • 2.
     Measures thestrength of correlation between a dichotomous (1,0) variable and a continuous variable (R).
  • 3.
    Dichotomous Variable Continuous Variable ExpectedResults Sex (male (1), Female (0)) IQ score Inferences about the relationship between sex and IQ Education ( private (1), public ((0) Math score Inference about the type of schooling and performance on Math Marital Status (Married (1), Single (0)) Cholesterol Level Inferences about marital status and health Location of residence (Urban(1), rural (0)) Level of literacy Inferences about location of residence and literacy College degree ( graduated (1), no college degree (0)) Income Inferences about level of education and income
  • 4.
     Measures thestrength of correlation between a dichotomous variable (1,0) and an ordinal variable (rank).
  • 5.
    Dichotomous Variable OrdinalVariable Expected results Sex (male 1, female 0) TV networks preferences Inferences about sex and preferred TV networks Education ( private 1, public 0) TV program preferences Inferences about type of schooling and preferred TV programs Marital Status ( married 1, single 0) Senatorial Candidate preferences Inferences about marital status and preferred senatorial candidate Location of residence (Urban 1, rural 0) Bb Pilipinas candidates preferences Inferences about location of residence and preferred BB. Pilpinas candidates
  • 6.
     Measures thestrength of correlation between a dichotomous variable (1,0) and another dichotomous variable (1,0).
  • 7.
    Dichotomous Variable Dichotomous Variable ExpectedResults Sex (male (1), Female (0)) College Education (Graduate 1, Non- graduate 0) Inferences about sex and educational attainment Education ( private (1), public ((0) Sports involvement (involved 1, not involved 0) Inference about the type of schooling and sports involvement Marital Status (Married (1), Single (0)) Financial status (Poor 1, not poor) Inferences about marital status and affluence Location of residence (Urban(1), rural (0)) Financial status (Poor 1, not poor 0) Inferences about location of residence and affluence College degree ( graduated (1), no college degree (0)) Employment (Employed 1, not employed 0) Inferences about level of education and employment
  • 8.
    Name X Ystatistic Point Biserial Coefficient Dichotomous variable Continuous Variable Rank Biserial Coefficient Dichotomous variable Ordinal Variable (Ranking) Phi coefficient Dichotomous variable Dichotomous variable pbr rbr
  • 9.
    Let X bea dichotomous variable and Y the continuous Variable  1 01 1 01     nn nn s MM r n pb       n i in YY n s 1 2 1 1 1 Sample standard deviation M1 – the mean value of observation in the continuous variable Y for group 1 M0 – the mean value of observation in the continuous variable Y for group 2 (0) n1 – size of sample for group 1 n0 – size of sample for group 2 (0) n – total sample size