News headline

1
PSY2010 Statistics for Experimental Psychology
News headline
A report by the Chief Inspector of Schools
has found that there is no link between class
size and educational attainment
Actual Report
Among children aged 5-7 there is some
evidence that class size has an effect,
however, for children aged 8 and over, there is
no statistically significant evidence of a
correlation between class size and educational
attainment
2
PSY2010 Statistics for Experimental Psychology
Correlation
Correlation measures the strength of a
relationship between two measurements on a scale
from -1 to 1. The values 1 and -1 relate to a
perfect linear relationship between the
measurements The value 0 relates to no
association between the measurements. Correlation
is only sensible when a scatterplot indicates a
linear trend.
3
PSY2010 Statistics for Experimental Psychology
4
PSY2010 Statistics for Experimental Psychology
Correlation
Calculations (X is IQ and Y is Grade Point
Average)
5
PSY2010 Statistics for Experimental Psychology
Correlation
Leading to


6
PSY2010 Statistics for Experimental Psychology
Correlation
A test of the significance of r is a test of the
null hypothesis of no association between the
variables X and Y
Refer to Table 6.2 in Neave with n12.
We have p lt 0.002 - very highly significant. We
reject the null hypothesis.
7
PSY2010 Statistics for Experimental Psychology
A word of caution
We must be very careful in interpreting
correlation coefficients. Just because two
variables are highly correlated does not mean
that one causes the other. In statistical
terms, we say that correlation does not imply
causation.
Examples X can cause Y (smoking and lung cancer)
Both X and Y respond to changes in some
unobserved variable (ice cream sales and shark
attacks!) Confounding. Need to design your
experiment appropriately to establish causation.
8
PSY2010 Statistics for Experimental Psychology
9
PSY2010 Statistics for Experimental Psychology
Spearmans rank correlation
Refer to Table 6.4 in Neave with n8.
Result is significant at 10 (0.05 lt p lt 0.1),
retain null but consider further investigation.
10
PSY2010 Statistics for Experimental Psychology
Kendalls rank correlation
21 concordant 7 disconcordant

• 14/28 0.5
• Use Table 6.5
• in Neave

11

• In an episode of The Simpsons (Season 7, "Much
  Apu About Nothing"). Springfield had just spent
  millions of dollars creating a highly
  sophisticated "Bear Patrol" in response to the
  sighting of a single bear the week before.
• Homer Not a bear in sight. The "Bear Patrol" is
  working like a charm!
• Lisa That's specious reasoning, Dad.
• Homer uncomprehendingly Thanks, honey.
• Lisa By your logic, I could claim that this rock
  keeps tigers away.
• Homer Hmm. How does it work?
• Lisa It doesn't work. (pause) It's just a stupid
  rock!
• Homer Uh-huh.
• Lisa But I don't see any tigers around, do you?
• Homer (pause) Lisa, I want to buy your rock.