Psy 231 Final Exam practice – Spring 2010
Note: Chits can used only on items not marked “NO CHITS.” Total chits: 10. Chits must be from the materials specified for the final exam.
Using appropriate symbols, write the definitional equation/formulae for the following statistics.
Note: write the complete formula as used in class, without using SS or other shorthand notations.
[NO CHITS]
(5) The arithmetic mean
(8) The sample standard deviation
(5) A standard (z) score
(5) The equation for a linear relation (use statistical symbols, not traditional mathematical symbols)
(3 each) Identify the three components of a good description of a set of data.
Note the relative strengths & weaknesses of each of the following research methods:
(4) Correlational method
(4) Experimental method
(4) Quasi-experimental method
(3 each)What are the two reasons that the normal distribution receives so much attention in statistics?
(3 each) What are the three most common measures of central tendency?
(6) Generally, why is the mean superior to the other measures of central tendency? And (3) what is the major weakness of the mean? [NO CHITS]
(5) Why is the standard deviation the most widely used descriptive measure of variability?
(3) When we compute the average, or mean, deviation score, what value always results?
(4) What causes this value to occur?
(4) How do we prevent the foregoing value occurring when computing standard deviation?
(4) According to Sir Karl Popper, what is the fundamental requirement of a scientific hypothesis?
(6) Explain the difference between hypothetical constructs and operational definitions. [NO CHITS]
(2 each) Name the four scales of measurement [NO CHITS]
(2 each) Provide the appropriate statistical label for each of the following symbols: [NO CHITS]
M ______s ______μ ______
Σ ______r______σ ______
(4 each) On the back of this sheet, name and very briefly explain the nine ethical principles as discussed in class that APA has established for research with adult humans. [NO CHITS]
(4 each) What are the two useful purposes that are served by transforming X-values into z-scores?
(4 each) Name and explain the two components of a z-score that describe the exact location of a score within a distribution.
(3) How does the numerical value of a z-score specify the distance of the score from the mean of a distribution?
(3 each) When every X value in a distribution is transformed into a corresponding z-score, the new z-score distribution has three properties. What are those three properties? [NO CHITS]
(3) Why do many people like to transform z-scores into a different standardized distribution?
(2 each) Name three common examples of such standardized distributions.
(4 each) Name and explain the three characteristics of a relation measured by a correlation. [NO CHITS]
(3 each) Sketch graphs depicting each of the following: [NO CHITS]
A positive relation A negative relation A curvilinear relation
(8) Name the two most commonly used correlations and explain how they differ.
(12) Sketch a graph of a normal distribution, mark the standard deviations on the x-axis, and note the areas (in %) under the normal curve for each of the following [Note: include two decimal places with your answers.] [NO CHITS]
0 – 1 s.d. ______, 1 – 2 s.d. ______, 2 – 3 s.d. ______, 3+ s.d. ______
Part 3: Computations - NO CHITS ON COMPUTATIONS
(6 each) For a sample with M = 40 and s = 8, find the following z-scores (show computations):
X = 47
X = 38
(12) Suppose your got a score of 90 on an English test for which M = 85, s = 8, and a score of 72 on a physics test for which M = 65, s = 6. On which test would you expect to receive a better grade? Show your computations and explain your answer.
Exam 2 # missed by Exam 1 # missed, 0-scores excluded
r = 0.64 r2 = 0.41 p 0.004 SEE = 10.7 Exam 2 # missed = 8.00 + .48(Exam 1 # missed)
How much of the variability in Exam 2 number missed can be explained by Exam 1 number missed?
What’s the likelihood of making a mistake if one says that Exam 2 number missed is moderately related to Exam 1 number missed?
On average, by how many points will you miss if you use the current regression line to estimate Exam 2 number missed based on Exam 1 number missed?
For the following by Exam 1 numbers missed, what is the Exam 2 number missed likely to be? Show your work!
# missed = 20
# missed = 65
What caution would you give to a person who wanted to know the estimated Exam 2 number missed for a person with 100 Exam 1 number missed based on this regression?
BONUS QUESTION: For Exam 1 number missed = 30, what is the probable highest Exam 2 number missed that a person might receive? What is the probable lowest Exam 2 number missed that a person might receive? Show your work.