January 2017 Answers
Question 1
A. Define reliability and discuss two forms of reliability where error variance due to time
sampling might affect the magnitude of the reliability coefficient
Reliability of a measure refers to the consistency with which it measures what it is supposed
to measure
Reliability = consistency of a measure
Where consistency lies there is always a certain amount of error
Two forms where error variance due to time sampling might affect the magnitude of the
reliability coefficient
1. Test-Retest Reliability
Administer it twice to the same set of group test-takers
Coefficient is called the coefficient of stability
Physical environment, testing conditions may contribute to systematic error variance
Transfer effects such as practice and memory play a role on second testing occasion
The magnitude of the reliability measure coefficient – is whether or not this is reliable
Standardized measures should be between .80 and .90
2. Alternative-Form Reliability
Two equivalent forms of the same measure are administered to the same group on two
separate occasions
The coefficient obtained between the sets of scores represent the reliability coefficient – also
called the coefficient of variance
This coefficient must range between .80 and .90 to be accepted
However the magnitude of the reliability coefficient can be effected
Because these measures have to be truly equivalent (should have the exact same number)
Similar to test-retest – transfer effects can contribute
B. Required magnitude of a reliability coefficient
Depending on what measures is used for depends on when the reliability coefficient is
acceptable
Standardized measures should have reliabilities between .80 and .90
One researcher (Huysamen) suggests that it should be higher than .85 for individuals and
higher than .65 for groups
Question 2
A psychologist assess two children experiencing the same difficulties. Both children are in Grade 1
and are experiencing difficulty in all academic areas. The first child’s age is 7 years, 2 months and the
second child’s age is 7 years, 3 months. Their teachers have complained that the children find it
difficult to remain focused for extended periods of time and that they are disruptive I the classroom.
They often interrupt the teacher and struggle to wait their turn. Both display very immature social
behaviour, are impulsive and often require adult support to resolve minor conflicts with peers in the
same grade. Both children have completed Grade R, but still do not know all the letters of the
alphabet.
Question 1
A. Define reliability and discuss two forms of reliability where error variance due to time
sampling might affect the magnitude of the reliability coefficient
Reliability of a measure refers to the consistency with which it measures what it is supposed
to measure
Reliability = consistency of a measure
Where consistency lies there is always a certain amount of error
Two forms where error variance due to time sampling might affect the magnitude of the
reliability coefficient
1. Test-Retest Reliability
Administer it twice to the same set of group test-takers
Coefficient is called the coefficient of stability
Physical environment, testing conditions may contribute to systematic error variance
Transfer effects such as practice and memory play a role on second testing occasion
The magnitude of the reliability measure coefficient – is whether or not this is reliable
Standardized measures should be between .80 and .90
2. Alternative-Form Reliability
Two equivalent forms of the same measure are administered to the same group on two
separate occasions
The coefficient obtained between the sets of scores represent the reliability coefficient – also
called the coefficient of variance
This coefficient must range between .80 and .90 to be accepted
However the magnitude of the reliability coefficient can be effected
Because these measures have to be truly equivalent (should have the exact same number)
Similar to test-retest – transfer effects can contribute
B. Required magnitude of a reliability coefficient
Depending on what measures is used for depends on when the reliability coefficient is
acceptable
Standardized measures should have reliabilities between .80 and .90
One researcher (Huysamen) suggests that it should be higher than .85 for individuals and
higher than .65 for groups
Question 2
A psychologist assess two children experiencing the same difficulties. Both children are in Grade 1
and are experiencing difficulty in all academic areas. The first child’s age is 7 years, 2 months and the
second child’s age is 7 years, 3 months. Their teachers have complained that the children find it
difficult to remain focused for extended periods of time and that they are disruptive I the classroom.
They often interrupt the teacher and struggle to wait their turn. Both display very immature social
behaviour, are impulsive and often require adult support to resolve minor conflicts with peers in the
same grade. Both children have completed Grade R, but still do not know all the letters of the
alphabet.
