The Simple Linear Regression Model
Regression in STATA
Survey data analysis
ANALYSIS OF SURVEY DATA (SPSS/STATA)
Kiplimo Araap Langat, PhD
Alexander Brookes Associates Limited,
The Old Church, Quicks Road Wimbledon, London, SW19 1EX, United Kingdom
trainer@alexanderbrookes.com
INTRODUCTION TO LINEAR REGRESSION & SURVEY DATA ANALYSIS
July 29, 2016
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Course Outline
1 The Simple Linear Regression Model
OLS
2 Regression in STATA
3 Survey data analysis
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
ANALYSIS OF SURVEY DATA (SPSS/STATA)
Kiplimo Araap Langat, PhD
Alexander Brookes Associates Limited,
The Old Church, Quicks Road Wimbledon, London, SW19 1EX, United Kingdom
trainer@alexanderbrookes.com
INTRODUCTION TO LINEAR REGRESSION & SURVEY DATA ANALYSIS
July 29, 2016
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Course Outline
1 The Simple Linear Regression Model
OLS
2 Regression in STATA
3 Survey data analysis
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Simple linear regression model
We can use various methods to derive estimates of the
parameters of economic relationships from statistical
observations.
Regression analysis is one of the most commonly used tools in
econometric work.
Regression analysis is concerned with describing and
evaluating the relationship between a given variable (the
explained or dependent variable) and one or more other
variables (often called explanatory or independent variables).
We will denote explained variable by y and explanatory
variables by x1,x2,.,xk.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
It has been noted above that there are various econometric
methods that can be used to derive estimates of the
parameters of economic relationships from statistical
observations.
The focus at the moment will be on the method of OLS or
classical least squares (CLS).
OLS has some attractive attributes that makes it popular.
1 Parameter estimates obtain by OLS has some
optimal/desirable properties.
2 Simple computational procedures
3 OLS has been used in a wide range of economic relationships
4 The mechanics of OLS are simple to understand
5 OLS is an essential component of most other econometric
techniques
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Regression line
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Fitting the regression line
Suppose there are n data points yi, xi, where i = 1, 2, , n. Such
that Y is a linear function of X , with unknown parameters α and
β. The goal is to estimate the values of these parameters in the
equation:
y = α + βx
Which would provide a best fit for the data points.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Suppose that we have a sample of 4 observations with X
values as shown.
If the relationship were an exact one, the observations would
lie on a straight line and we would have no trouble obtaining
accurate estimates of α and β.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
In practice most economic relationships are not exact and the
actual values of Y are different from those corresponding to the
straight line.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
To allow for such divergence, we will write the model as
y = α + βx + µ
Where µ is a disturbance term:
Therefore, each can be expressed in terms of two components,
one component due to Xi and a second component due to the
influence included in the random term µi .
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Each value of Y thus has a non-random component, α + βx, and
a random component, µ.
The first observation has been decomposed into these two
components.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
In reality we can only see the P points
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
We can use the P points to draw a line which is an approximation
to the line
y = α + βx
If we write this line
yb = a + bx
Where a is an estimate of α and b an estimate of β.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The line is called the fitted model and values of Y predicted by it
are called the fitted values of Y. they are given by the heights of R
points. Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The fitted line values are given by the heights of R points.
The P points represent the Y actual value while the fitted values
represented by R points and the heights
Kiplimo Araap Langat, PhD
of the R pints.
ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The discrepancies between the actual and fitted values of Y are
known as the residuals e.
e(residual) = Y − yb
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
NB/ the values of the residuals are not the same as the values of
the disturbance term. The diagram below now shows the true
unknown relationship as well as fitted line.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The disturbance term in each observation is responsible for
divergence between the non-random component of the true
relationship and the actual observation.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The residuals are the discrepancies between actual and the fitted
values.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
If the fit is a good one, the residuals and the values of the
disturbance term will be similar, but they must be kept apart
conceptually.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Decomposition
Both of these lines will be used in our analysis.
Each permits a decomposition of the value of Y.
The decomposition will be illustrated with the fourth
observation.
Using the theoretical relationship, Y can be decomposed into
its non-stochastic component α + βx and its random
component µ.
This is a theoretical decomposition because we do not know
the values of α or β, or the values of disturbance term µ.
We shall use it in our analysis of the properties of regression
coefficients.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The other decomposition is with reference to the fitted line.
In each observation, the actual value of Y is equal to the
fitted.
In each observation, the actual value of Y is equal to the
fitted value plus the residual.
This is an operational decomposition which we will use for
practical purposes.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Assumptions of Linear Stochastic Regression Model
Linear regression model is based on certain assumptions, some of
which refer to the distribution of the random variable µ, some to
the relationship between µ and explanatory variables X.
1 Normal distribution;
2 Zero Expected value/mean;
3 Constant variance:
The first three assumptions can be summarized as
µ ∼ N(0, σµ2 ) (1)
4 Uncorrelated error terms; E (µi µj ) = 0
5 Explanatory and error terms are uncorrelated; E (Xi µi ) = 0
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Assumption ]1: Normal distribution
The first assumption of the Classical Linear Regression model
(OLS) is that random error term µ is normally distributed.
As a result, Y and the sampling distribution of the parameters
αandβ of regression are also normally distributed.
This is necessary so that tests can be conducted on the
significance of parameters.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Assumption ]2: Zero Expected value/mean
The second assumption is that the expected value of the error
term or its mean is equal to zero.
E (µi )2 = 0
Specifically; since X is assumed to be fixed, the value of Y
varies above and below its mean as µ exceeds or is smaller
than 0.
Therefore since µ is assumed to be 0, the given yb gives the
average value of Y.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Assumption ]3: Constant variance
the third assumption is that the variance of the error term is
constant in each period and for all values of X; that is,
E (µi )2 = σµ2
This assumption ensures that each observation is equally
reliable, so that estimates of the regression coefficients are
efficient and tests of hypothesis about them are not biased.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
These first three assumptions about the error term can be
summarized as
µ ∼ N(0, σµ2 )
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Assumption ] 4: Uncorrelated error terms
The fourth assumption is that the value with the error term
assumes in one period is uncorrelated or unrelated to its value
in any other period; that is,
E (µi µj = 0
for i 6= j; i, j = 1, 2, , n
This ensures that the average value of Y depends only on X
and not on µ , and it is once again required in order to have
efficient estimates of the regression coefficients and unbiased
tests of their significance.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Assumption ]5: Explanatory and error terms are
uncorrelated
The fifth assumption is that the explanatory variable assumes
fixed values that can be obtained in repeated samples, so that
the explanatory variable is also uncorrelated with the error
term; that is,
E (Xi µi ) = 0
This assumption is made to simplify the analysis.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Least Square Criterion
The OLS method gives the best straight line that fits the
sample of XY observations in the sense that it minimizes the
sum of squared (vertical) deviations of each observed point on
the graph from the straight line.
We take vertical deviations because we are trying to explain or
predict movements in Y, which is measured along the vertical
axis.
We cannot take the sum of the deviations of each of the
observed points from the OLS line because deviations that
equal in size but opposite in sign cancel out, so the sum of
the deviations equals zero.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The student’s t Test and F-test
To test the significance of parameter estimates we use the
students t-test
To test the overall fit of the model we use F-test.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Importance of the Statistical Tests of Significance
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Summary of OLS Procedure for a two Variable Model
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Properties of OLS
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Properties of OLS
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Properties of OLS
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Properties of OLS
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Properties of OLS
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Violations of the OLS
OLS Assumptions implied that
Errors µ are Normally Distributed with mean 0
Errors have constant variance σ 2
Errors are independent
X is Measured without error
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Violations of the OLS
OLS Assumptions implied that
Errors µ are Normally Distributed with mean 0
Errors have constant variance σ 2
Errors are independent
X is Measured without error
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Violations of the OLS
OLS Assumptions implied that
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Non-Normal errors
OLS violation
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Non-Constant Error Variance (Heteroscedasticity)
OLS violation
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Measurement Error in X
OLS violation
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Course Outline
1 The Simple Linear Regression Model
OLS
2 Regression in STATA
3 Survey data analysis
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Regression Analysis
Stata is capable of many types of regression analysis and
statistical tests.
However, in this section, we touch on only a few of the more
common commands and procedures.
The case of limited dependent variable will be analysed in the
following chapter:
We consider the following relevant commands:
regress
predict
test,
testparm
ovtest
hettest
svy option
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
regress
regress command carries out a regression analysis on the
variables specified.
Regression analysis estimates an equation of independent (or
explanatory) variables that best explains variation in a
dependent variable.
This command is used when the dependent variable is a
continuous variable.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Syntax
The syntax is:
regress DV IV(s) [if exp] [in range] [options]
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Stored scalars
Regress stores the following scalars among other in e():
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Post-estimation in reg command
The following are some of the postestimation commands of special
interest after regress:
estat (svy) is the postestimation statistics for survey data
1
1
Click here to read more
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
predict command
predict is a post estimation predictions of residuals, influence
statistics, and other diagnostic measures
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Practical STATA session in class on regression
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Course Outline
1 The Simple Linear Regression Model
OLS
2 Regression in STATA
3 Survey data analysis
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Survey data analysis
The fact that survey data are obtained from units selected
with complex sample designs needs to be taken into account
in the survey analysis;
Weights need to be used in analysing survey data and
variances of survey estimates need to be computed in a
manner that reflects the complex sample design.
Most data are collected through surveys with
probability-based complex sample.
Survey data are collected under different sampling plans will
require special treatment to account for complex nature of
data in the analysis.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Example
Suppose there are 200,000 households in the population, of
which 160,000 live in rural areas.
A survey is carried out with a random sample of 2000 urban
households and 2000 rural households.
Thus, the population is 80% rural but the sample is only 50%
rural.
Suppose per capita income is 2800 in urban areas and 1400 in
rural areas.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Example
The simple average income across households in the sample
would be 2100, but we know that the average income across
the country would be lower than this because rural households
(who have lower incomes) are under-represented in the sample.
We can compensate for this by using a weighted average, with
weights equal to the ratio of the population to the sample.
The urban weight would be
20(= 40 thousand/2 thousand)and the rural weight would
be 80(= 160 thousand/2 thousand).
The weighted average is calculated as
(80∗1400+20∗2800)/(20+80) = 0.8∗1400+0.2∗2800 = 1680
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The complex nature of these sample designs causes a
departure from the assumption of independent sample points
having equal probabilities of selection.
This departure creates a requirement for specialized statistical
software to accurately compute estimates of population
statistics and their standard errors.
Otherwise, the standard errors produced, as for a simple
random sample, would generally underestimate the true
population value.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Survey data analysis
The complex design of sample surveys dictate that the
statistical analysis procedures used to analyse the data have
the ability to account for:
multiple stages of sampling,
stratification, and
clustering.
The accompanying SE must be computed using procedures
that take into account the complex nature of the sample
design.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Effects of sampling design
Declaring the data to be survey
sveyset data consider the sampling design
Most surveys are based on stratified cluster samples rather
than pure random samples.
Sample design affects the calculation of averages and
percentages, so we need to calculate weighted averages and
percentages to compensate for the fact that some households
are over-represented in the sample, while others are
under-represented.
The sample design also affects the calculation of standard
errors in regression analysis.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Sampling designs and their effects
1 Stratification: The goal of stratification is to over-represent
groups of households that are highly diverse in the variables of
interest (e.g. income). If well done, stratification therefore
increases the accuracy of estimates (that is, it reduces the
standard errors) compared to a simple random sample.
2 Clustering: The goal of using clusters of households in
samples is to reduce the cost of data collection, but this
reduces the accuracy of estimates (that is, it increases the
standard error) compared to a non-clustered random sample.
To see this, imagine the difference between interviewing 100
households dispersed across the country and interviewing 100
households in one village. Clearly, estimates based on the
latter would be less accurate.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Declaring survey data
The svyset command is used to describe the sample design.
Then the svy: prefix is used before other commands such as
regress and probit.
syntax for svyset is organized according to each level in the
sample design.
First define the primary sampling unit PSUs
Next, we define the strata
Third, in the svyset command, we specify the primary
sampling unit variable (psu), and the sampling weight variable
(weight), the strata variable (sector) and clusters if given
The two vertical lines followed by n indicate that in the second
stage, the sampling was random.
svyset command declare survey design for dataset
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Syntax for survey data from a one-stage design
A commonly used single-stage survey design uses clustered
sampling across several strata, where the clusters are sampled
without replacement. In a Stata dataset composed of survey
data from this design, the survey design variables identify
information about the strata, PSUs (clusters), sampling
weights, and finite population correction.
Here we use svyset to specify these variables, respectively
named strata, su1, pw, and fpc1.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
Syntax
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
References
Main Text
Salvatore, D. and D. Reagle (2001). Statistics and Econometrics, Schaums
Outline Series. 2ed. McGraw-Hill, New York.
Other Texts
Mukras, M.S., (1993). Elementary Econometrics: Theory, Application and Policy.
EAEP. Nairobi.
Madala, G.S. (2001). Introduction to Econometrics. John Willey and Sons,
England.
Koutsoyiannis, A. (1977).Theory of Econometrics. 2ed, ELBS, Ontario.
Durlauf, S. N., and Blume, L. (Eds.). (2008). The new Palgrave dictionary of
economics (Vol. 6). Basingstoke: Palgrave Macmillan.
Granger, C. W. (2004). Time series analysis, cointegration, and applications.
American Economic Review, 421-425.
Long, J. S., and Trivedi, P. K. (1992). Some specification tests for the linear
regression model. Sociological Methods and Research, 21(2), 161-204.
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
The Simple Linear Regression Model
Regression in STATA
Survey data analysis
The End
Kiplimo Araap Langat, PhD ANALYSIS OF SURVEY DATA (SPSS/STATA)
