PS11A Psychology Laboratory

1
PS11A Psychology Laboratory Methods Abnormal,
Social Developmental

• Charlene Coore
• 2006-2007

2
Reliability

• Reliability The extent to which a measurement
  instrument yields consistent results when the
  characteristic being measured hasnt been
  changed.

3
Reliability (Cont/d)

• Interrater reliability is the extent to which 2
  or more individuals evaluating the same product
  or performance give identical judgements.
• Internal consistency reliability is the extent
  to which all the items within a single instrument
  yield similar results.

4
Reliability (Cont/d)

• Equivalent forms of reliability is the extent
  to which two different versions of the same
  instrument yield similar results.
• Test-retest reliability is the extent to which
  the same instrument yields the same result on two
  different occasions.

5
Validity

• Validity is the extent to which a measurement
  technique measures what it claims to measure.

6
Internal Validity

• Internal Validity is the extent to which the
  design of a study ensures that the Independent
  variable, and not some variable or variables,
  caused the difference in the dependent variables.
• A Confound is an uncontrolled extraneous variable
  or another flaw in the research design that
  yields alternative explanations for the results
  and thus limits the studys internal validity

7
Examples of Confounds

• Experimenter Effects
• Experimenter Bias
• Demand Characteristics

8
External Validity

• External Validity is the extent to which the
  results of an investigation can be generalised to
  beyond study.

9
External Validity Sampling

• Population
• Sample
• Random Selection
• Convenience Sample
• Random Assignment

10
Validity (Cont/d)

• Face Validity is the extent to which, on the
  surface, an instrument looks like its measuring
  a particular characteristic
• Content Validity is the extent to which a
  measurement instrument is a representative sample
  of the content area/domain being studied

11
Validity (Cont/d)

• Criterion Validity is the extent to which the
  results of an assessment instrument correlate
  with another.
• Construct validity is the extent to which an
  instrument measures a characteristic that cannot
  directly be observed but must instead be inferred
  from patterns in peoples behaviour (e.g.
  motivation, creativity, racial bias).  

12
Hypotheses

• Hypothesis a possible answer to a research
  question
• Research Question Are students in Social
  Sciences (Soc Sci) smarter than students in other
  faculties?
• Hypothesis The IQ scores of students in Soc Sci
  will be higher than those of the university
  students in other faculties.

13
The Null Hypothesis (Ho)

• Null Hypothesis the prediction that there is no
  difference between groups being compared. The
  null hypothesis is true if the population that
  the sample is from is the same as the population
  with which it is being compared.

14
The Alternative Hypothesis (Ha/H1)

• Alternative Hypothesis the prediction that the
  researcher is making about the results of
  research the researcher predicts that there will
  be a difference between the groups being compared.

15
Alternative Hypothesis

• There are 2 types of Alternative Hypotheses. 
• Two-tailed Hypothesis a type of alternative
  hypothesis in which the researcher simply
  predicts that the two groups being compared will
  differ, but does not predict the direction of
  that difference.
• One-tailed Hypothesis a type of alternative
  hypothesis in which the researcher predicts the
  direction of the difference between the groups
  being compared.

16
Hypothesis Testing

• Hypothesis Testing the process of determining
  if the hypothesis is supported by the results of
  a research project.
• Reject the Ho vs. Fail to reject the Ho

17
Errors in Hypothesis Testing

• Hello were only human!
• Sometimes we reject the null hypothesis when it
  is really true.
• OR we fail to reject the Ho when we should reject
  it

18
The Four Possible Research Outcomes
The Truth
19
Type I Type II Errors

• Two ways to be correct
• Reject Ho when it is in fact false.
• Fail to Reject Ho when it is true.
• Two possible errors
• Reject the Ho when it is true (Type I error)
• Fail to reject Ho when it is false (Type II
  error)

20
Which Error is more Serious?
The Truth