Answer all of the following questions. For every question, please state the type of inferential test used (e.g.,
paired samples t-test, 2 x 4 factorial ANOVA, etc.), show the obtained value of the test statistic(s), e.g the F-value,
the t-value, etc., the
1.
(10 points) A Rutgers alumnus (summa cum laude in economics) hypothesizes that graduates of RU on average
earn higher annual salaries in the first year after graduation than do first year graduates of Seton Hall. A
random sample of 65 1st year graduate
Multiple Regression Analysis:
First we calculate the Pearson's Rs between the predictors and the criterion. If they are both significant, we include them in the
equation. If one is not significantly correlated (and the other is), we do simple linear reg
Review
True or False Questions at the end of the Chapter on Pearson's r (hardcover, 329)
22) Significant correlations always predict better than chance. True
23) Negative correlations, even when significant, never predict better than chance. False
24) Th
Pearsons r The Correlation Coefficient
When both the IV (X) and DV (Y) are
interval/ratio level, r is a measure of the strength
and direction of their association
There are two things you need to describe a straight line: The slope (and the direction bei
Final Exam Review, Part 2
A restaurant chain wants to see if serving appetizers that are higher in sodium increases the number of drinks customers order.
Randomly assigned customers are given low-sodium (n = 10), moderate sodium (n = 10), or high sodium
Final Exam, Part 3
(Page 422, softcover; Problem 17):
A researcher is interested in the possible intercorrelations among three of the subtest scores on the Wechsler Adult Intellignece
Scale (WAIS). A random sample of 10 inmates from a Federal Prison were
Final Review, Part 1
A scale of self-esteem (where 1 = very low self-esteem and 10 = very high self-esteem) has been developed with a known
population mean of 5.0 and a known population standard deviation of 2. A sample of people diagnosed as depressed (
Let's say College GPA is correlated with HS GPA with a correlation of 0.6. This accounts for 36% of the variability. Now let's
say SAT scores account for another 36% of College GPA. Notice the overlap. SAT and HS GPA are closely correlated.
The HS
End of Chapter Conceptual Questions on the Paired-Samples T-test: (Paperback, 453; Hardcover, 475)
10) For a given dierence between the means and equal numbers of scores, which test, the paired t or the independent t, has
more statistical power? (For thi
How many levels of the independent variable in a one-way ANOVA? More than two.
Remember, every time you do a t-test, you increase the probability of committing a type 1 error
Skip this page (Steps 1-4) about calculating the sum of squares between and wi
Exam 1 Review
Practice Problems from last semesters test:
1) A scale of self-esteem (where 1 = very low self-esteem and 10 = very high self-esteem) has been developed with a known
population mean of 5.- and a known population standard deviation of 2. A s
Review
Independent Samples T-Test
We do this when we are asked if the dierence between two sample means is due to chance. The sample means are point
estimates of population means. It is the sampling distribution of the dierence (not of the mean). It is a
End of Chapter 4 Problems (Page 103, hardcover; Page 97, paperback.)
12) On a normal distribution, at what percentile must:
a) the mean fall? 50th
b) the median fall? 50th
c) the mode fall? 50th
If they don't fall at the 50th percentile, it is not n
Article Newspaper Discussion:
There is no such thing as "statistically insignificant."
A confidence interval is found in the line: "well within the poll's margin
of 4.5 percentage points." This is one example of its use. However,
they do not provide if it
End of Chapter True or False Problems 24-30; 35; 37-42 (Answers Only)
Page 211 in paperback edition. Page 243 in the hardcover edition.
(24) True. (25) False. Simply because two variables occur together, it does not mean one causes the other. Correlation
Review:
A Z score of 1.96 allows us to be 95% sure of the probability. A Z score of 2.58 allows us to be 99% sure of the probability.
Practice Problem 1: (Page 131, Paperback; 167, Hardcover)
The problem is asking us for the probability of scoring a 52 o
How to Calculate the Mean and Standard Deviation in Microsoft Excel (2010)
Place your values for a data set in Column A. Highlight the first empty cell under the column in question. In the tool bar:
Mean: Formulas -> Math and Trig. Functions -> Average -
Fundamental Principles and Basic Terminology
(Covered in Chapter 9 of the textbook; You can ignore Chapter 1 entirely)
Measurement - The heart of any scientific endeavor. Science requires quantitative data obtained
through the measurement of data.
Relia
Review
If we are asked to calculate a confidence interval (let's say 95%) and we are given the population standard deviation and the
sample mean (the population mean is unknown):
If we don't have the population SD and we only have the sample SD:
What are
Review
When do we use the z-test? When we have a sample mean and the population Standard Deviation and we are being asked to
compare it to the population mean.
When do we use the t-test? When we have a sample mean and we want to see if it comes from a p
Recitation 8
3/30/17
Tukeys HSD Test
HSD = F-critical value at 0.05 or 0.01 (Table H) * (MSw/n)
n=number of observations/values per group
DF used (within degrees of freedom (on left side) and number of
sample groups across the top row).
Factorial ANOVA
Recitation #1
1/26/2017
Introduction
Email: [email protected]
Office: Tillett Hall Room 409
Office Hours: After recitation or by appointment
*This information can be found
on the course syllabus
Recitation: What to expect
Recitation will include
Recitation 3 & 4
2/16/2017
Chapter 5: Sampling
Distribution
Sampling distribution-A frequency distribution comprised of sample
means.
Mean of these means can be considered the population mean.
Using a z-test we can determine the probability that a given
Recitation 2
2/2/2017
Note
Please let me know about any inconsistencies between
lecture and recitation.
For example, in the SD equation x=(X-M)
Kurtosis
SD of normal deviation tends to be 1/6th of its range
Mesokurtic
SD is 1/6th of its distribution (
Recitation 5
2/23/2017
Chapter 11/13: Chi Square
Comparing the observed data to the expected data
fo = the frequency observed
fe = the frequency expected
We compare the observed data to the expected data to
determine how well the observed data fit wha
Chapter 11: The Correlation Coefficient Pearsons r
Pages. 287-296
Cause and Effect pg. 287
Epidemiological studies (post-facto research): based on observing groups of people and then
showing statistical correlations between their lifestyles or behavior an