STA 6126 Statistical Methods in Social Research I
Spring 2014
Homework #4
Assigned Friday February 7
Due Friday February 14
Reading Lecture notes pp. 79108
Problems on Basic Probability (Use set notation; cite probability rules.)
1 All human blood can be
Statistics 6126
Brief Solutions to Homework Exercises
These solutions are solely for the use of students in STA 6126 at the University of Florida and are not to
be distributed elsewhere. Please report any errors in these solutions to Alan Agresti, aa@stat
Practice questions: EXAM 2
STA 6126
1. A study was conducted of the eects of a special class designed to aid students
with verbal skills. Each child was given a verbal skills test twice, both before
and after completing a 4-week period in the class. Let Y
x
y E d D U I G a ` V d U E a ` V ` i E Y
We 3 RrvbiFqPou Tu p RnIufFrvRbIXrbGWPXWS
EWufXbEWbtXpRbtVyPFoeFXuvPFWqXPE jghTf
DV V d DV `tt I S ` ` p E i I Q E DV IV i E i i d EY i g f q
d
cmk F
V U IV `Y S E DV IV tY Eq EY a `Vt E U k E dq Y I Ck E UYV
STA 6126 Statistical Methods in Social Research I
Spring 2014
Homework 1
Assigned Wednesday January 8
Due Wednesday January 15
Note: AF refers to the textbook by Agresti and Finlay.
Reading Handout from Day 1 (Ch. 1 of Statistics, by Freedman, Pisani and
STA 6126 Statistical Methods in Social Research I
Spring 2014
Homework 2
Assigned Friday January 17
Due Wednesday January 29
PARTIAL SOLUTIONS January 31
Reading AF (textbook) Chapter 3
Exercises to hand in
1 Download R from http:/www.r-project.org and in
STA 6126 Statistical Methods in Social Research I
Spring 2014
Homework 3
Assigned Friday January 31
Due Friday February 7
Reading Read Chapter 4; see suggestions below.
4.3 The normal distribution; read all.
4.1, 4.2 Probability basics; lecture notes will
Name:
STA 6126 Statistical Methods in Social Research I
Spring 2014
Midterm 1
February 17, 2014
INSTRUCTIONS:
1. Please print your name in the space above. If you separate out the pages, then print your name on all
the pages.
2. Do the work on the exam. U
STA 6126 Statistical Methods in Social Research I Spring 2014
Sample Midterm 1 Questions
SOLUTIONS
1. Infant mortality rates are reported by the Centers for Disease Control as number of infants who die in the
rst year of life, per 1000 live births. The CD
Table 1: Table for Exercise 1.
VICTIMS
Frequency
Percent
-0
1244
90.8
1
81
5.9
2
27
2.0
3
11
0.8
4
4
0.3
5
2
0.1
6
1
0.1
N
Mean
Std Dev
-1370
0.146
0.546
PRACTICE QUESTIONS FOR EXAM 1
1. The 1990 General Social Survey asked respondents, During the past 12
FORMULAS
Bivariate regression models
E (y ) = + x
b t(se) t =
y = a + bx
r = b(sx/sy )
b
(df = n 2), se =
se
r2 = (T SS SSE )/(T SS )
s
s
s=
=
sx n 1
(x x)2
S SE
(n 2
s = S SE/(n 2) = Root MSE
Multiple regression models
E (y ) = + 1 x1 + 2 x2 + + k xk
R2
STA 6126 Exam 2
1.
Based on responses of 1600 subjects in General Social Surveys in the mid-1980s, a
95% Confidence Interval for the mean number of close friends equals (5.6, 7.2). For
each statement, circle the best response: Correct or Incorrect.
a)
b)
STA 6126 Exam 2
1.
Based on responses of 1600 subjects in General Social Surveys in the mid-1980s, a
95% Confidence Interval for the mean number of close friends equals (5.6, 7.2). For
each statement, circle the best response: Correct or Incorrect.
a)
b)
Some formulae you may need
Estimate of SE(Estimator)
Parameter
Estimator
X
X
S/ n
Xd = d
Sd / n
d = 1 2
(Dependent Samples)
1 2
(Independent Samples)
2
1
2
2
1 2
(Independent Samples)
=
2
1
2
2
2
1
Test Statistic and its distribution
X 0
T=
~ t( n 1 )
For each combination of response and predictor,
a) What do we want to infer?
b) Which test(s) should we use?
c) What are the assumptions needed?
(Factor)CategoricalPredictor is
Response has a
Normal
Distribution
Response is
Categorical
One Group
Inference
STA 6126 Spring 2012 Project 1
Due: Monday, March 28, 2012 at 10:40
Late reports will not be accepted.
Analysis of House Sales Data for Gainesville
1. Summarize the data below using your favorite statistical package, by means of one
table and one graph fo
STA6126 Announcements Spring 2012
Look at this page every day for news, links, etc.
Wednesday, March 14, 2012
Please check the documents folder for some formulas to use in Test-2, practice problems for
confidence intervals and hypotheses testing and Proj
Formulas yi n (y y )2 n1
y=
s2 = z=
y = n y t(se), t= y 0 se
y se = s/ n
=
(1 ) n se = (1 ) n 0 (1 0 ) n
z (se), z= n= n= 0 se0 z M z M
2
se0 = (1 )
2
2
1
Associations Between
Categorical Variables
Case where both explanatory (independent)
variable and response (dependent) variable
are qualitative (Chapter 7 includes case
where both are binary (2 levels)
Association: The distributions of responses
differ
STA 6126 Statistical Methods in Social Research I
Spring 2014
Homework #6
Assigned Monday March 10
Due Monday March 17
Reading
Chapter 6, pp. 169-171 on The Binomial Distribution
Chapter 4, Section 5 Central Limit Theorem
Chapter 5, pp. 107-111 up to Larg
STA 6126 Statistical Methods in Social Research I
Spring 2014
Homework #5
Assigned Friday February 21
Due Friday February 28
Reading Note on order of reading: We will talk about the binomial distribution before we talk
about the Central Limit Theorem in c
A multiple regression model is fit, relating salary (Y) to the following predictor variables:
experience (X1, in years), accounts in charge of (X2) and gender (X3=1 if female, 0 if
male). The following ANOVA table and output gives the results for fitting
STA 6126 Exam 1 Fall 2004
1. A questionaire is sent out to academic social scientists who are members of a national
association. For each of the following elements of the survey, are the responses best
described as Interval, Nominal, or Ordinal scale?
a)
STA 6126 Fall 2006 Exam 2
PRINT Name _
A research article conducts a significance test to determine whether the mean food
expenditures differ among families in two societies. Their 95% Confidence Interval
for is (32 , 48). Which of these statements best d
STA 6126 Exam 1 Fall 2005
1. A questionaire is sent out to academic social scientists who are members of a national
association. For each of the following elements of the survey, are the responses best
described as Interval, Nominal, or Ordinal scale?
a)
STA 6126 Fall 2007 Exam 4
PRINT NAME _
True or False?
1. The correlation coefficient (r) is a more appropriate measure to report than the
slope of the regression line (b) when describing the association between two
variables when there is not a clear inde
STA 6126 Exam3 Fall 2004
PRINT NAME_
1. Interchanging two rows in a contingency table will not have an effect on gamma.
True or False.
2. Interchanging two rows in a contingency table will not have an effect on the chisquared statistic. True or False.
3.
STA 6126 Exam 3 Fall 2005 PRINT Name _
1. Interchanging the rows (levels of the independent variable) in a contingency table
will:
a)
b)
c)
d)
Effect the chi-square statistic, but not the numbers of concordant/discordant pairs
Effect the numbers of concor