Psychology and Statistics

1
Psychology and Statistics

• Interpreting Data
• (Ch. 1 Myers and Ch. 2 Barrons)

2
Descriptive Statistics

• Descriptive statistics describes a set of data
• Measures of Central Tendency
• Goal is to determine the center of the
  distribution
• Measures of Variation
• Goal is to describe the diversity of the
  distribution

3
Measures of Central Tendency

• MEAN - Average
• MEDIAN Middle, 50th
• MODE Most frequently occurring score

4
Mean

• Most frequently used measure of central tendency
  but can be distorted by outliers (extremes). If
  the distribution includes outliers, then the
  median is a better measure of central tendency.
• Ex. 2, 7, 8, 9, 10, 18

5
Mean

• If the distribution is not skewed it is called
  symmetrical. Outliers skew the distribution. If
  there are high positive outliers then the
  distribution is positively skewed. (more low
  scores than high scores) If there are very low
  outliers then the distribution is negatively
  skewed. (more high scores than low scores)

Symmetrical Positively Skewed
Negatively Skewed
6
Examples of Skewed Distributions

• 39,000 penniless people
• Geography Majors from UNC Chapel Hill class of
  84??
• Test average 65

7
Measures of Variation

• Range the gap between the lowest and the
  highest score

8
Distribution of ValueStandard Deviation

• THE BELL CURVE (IQ)
• or 1 Standard Deviation 15 pts.
• Remember Standard Deviation measures the distance
  from the mean.
• Mean of IQ 100
• What score is 1 SD? What score is 1 SD?
• and 1 SD of IQ scores includes 68
• What score is 2 SD? What score is 2 SD?
• and 2 SD of IQ scores includes 95

9
Z Scores and Standard Deviation

• Z scores measure the distance of a score from the
  mean in units of standard deviation. If the score
  is below the mean its z score is negative. If it
  Is above the mean then the z score is positive.
• Consider this example. Mean of 100. Joe scores 90
  and has a z score of -1. Sarah scores a 105. What
  is her z score?

10
Inferential Statistics

• Inferential Statistics determines if the findings
  can be applied to a larger population from which
  the sample was selected.
• It is a goal of the researchers for the sample to
  be representative of the population
• The extent to which the sample differs from the
  population is known as sampling error.
• Statistical tests yield p values. The smaller the
  p-value the more significant the results. The
  cutoff for findings being statistically
  significant is p .05. This means that there is
  a 5 chance that the findings were by chance.
  P-value never equals 0 because we can never be
  100 certain that our findings were not due to
  chance.

11
Statistical Significance

• When is the difference significant?
• Psychologists are conservative when it comes to
  declaring some research statistically
  significant.
• Statistical significance When the difference
  observed is not likely to have occurred by
  chance.
• When samples averages are reliable and the
  difference between them is large then the
  difference has statistical significance
• When it is 5 or less likely to have occurred by
  chance, then psychologist will say that there is
  statistical significance.
• Be wary of statistical significance in
  homogeneous samples (these samples are
  homogeneous and not practical to life application)

12
Statistics can Lie

• Examples of Statistics being used for selfish
  purposes of politicians and/or corporations.
• Video
• ECU study changing NC school start times
• Global Warming Studies funded by Exxon.
• NC HWY 74 By Pass studies and impact on the
  environment.
• Health Care Debate - Lies, Lies and Statistics