Spss COMPLEX SAMPLES 13.0 User Manual

Download

Page 1

SPSS Complex Samples™ 13.0

Page 2

For more information about SPSS®software products, please visit our Web site at http://www.spss.com or contact

SPSS Inc. 233 South Wacker Drive, 11th Floor Chicago, IL 60606-6412 Tel: (312) 651 Fax: (312) 651-3668

SPSS is a registered trademark and t he other product names are the t rademarks of SPSS Inc. for its proprietary computer software. No m the trademark and license rights in the software and the copyrights in the published materials.

The SOFTWARE and documentation are provided with RESTRICTED R IGHTS. Use, duplication , or disclosure by the Governmentis clause at 52.227-7013. Contractor/manufacturer is SPSS Inc., 233 South Wacker Drive, 11th Floor, Chicago, IL 60606-6412.

General notice: Other product names mentioned herein are used for identification purposes only and may be trademarks of their respec

TableLook is a trademark of SPSS Inc. Windows is a registered trademark of Microsoft Corporation. DataDirect, D Portions of this product were created using LEADTOOLS © 1991–2000, LEAD Technologies, Inc. ALL RIGHTS RESERVED. LEAD, LEADTOOLS, and LEADVIEW are registered trademarks of LEAD Technologies, Inc. Sax Basic is a t All rights reserved. Portions of this product were based on the work of the FreeType Team (http://www.freetype.org). A portion of th software is provided “as is,” without express or implied warranty. A portion of the SPSS software contains Sun Java Runtime libraries. Copyright © 2003 by Sun Microsystems, Inc. All rights reserved. The licensed from IBM and are available at http://oss.software.ibm.com/icu4j/.

No part of this p electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publisher.

1234567890 060504

ISBN 1-56827-

-3000

aterial describing such software may be produced or distributed without the written permission of the owners of

subject to restrictions as set forth in subdivision (c) (1) (ii) of The Rights in Technical Data and Computer Software

tive companies.

ataDirect Connect, INTERSOLV, and SequeLink are registered trademarks of DataDirect Technologies.

Sun Java Runtime libraries include code licensed from RSA Security, Inc. Some portions of the libraries are

04 by SPSS Inc.

ublication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,

353-3

Page 3

Preface

SPSS 13.0 is a comprehensive system for analyzing data. The Complex Samples optional a manual. The Complex Samples add-on module must be used with the SPSS 13.0 Basesystemandiscompletelyintegratedintothatsystem.

Installation

To install the Complex Samples add-on module, run the License Authorization Wizard using the authorization code that you received from SPSS Inc. For more informat

Compatibility

SPSS is designed to run on many computer systems. See the installation instructions that cam requirements.

dd-on module provides the additional analytic techniques described in this

ion, see the installation instructions supplied with the SPSS Base system.

e with your system for specific information on minimum and recommended

Serial N

Your ser this serial number when you contact SPSS Inc. for information regarding support, payment, or an upgraded system. The serial number was provided with your Base system

Customer Service

If you have any questions concerning your shipment or account, contact your local office your serial number ready for identification.

umbers

ial number is your identification number with SPSS Inc. You will need

, listed on the SPSS Web site at http://www.spss.com/worldwide. Please have

iii

Page 4

Training Seminars

SPSS Inc. provides both public and onsite training seminars. All seminars feature hands-on workshops. Seminars will be offered in major cities on a regular basis. For more informa

tion on these seminars, contact your local office, listed on the SPSS

Web site at http://www.spss.com/worldwide.

Technical S

The service

upport

s of SPSS Technical Support are available to registered customers. Customers may contact Technical Support for assistance in using SPSS or for installation help for one of the supported hardware environments. To reach Technical Support, s

ee the SPSS Web site at http://www.spss.com, or contact your local office, listedontheSPSSWebsiteathttp://www.spss.com/worldwide.Bepreparedto identify yourself, your organization, and the serial number of your system.

Additional Publications

Additional copies of SPSS product manuals may be purchased directly from SPSS Inc. Visit the SPSS Web Store at http://www.spss.com/estore, or contact your local SPSS offi

ce, listed on the SPSS Web site at http://www.spss.com/worldwide.For telephone orders in the United States and Canada, call SPSS Inc. at 800-543-2185. For telephone orders outside of North America, contact your local office, listed on the SP

SS Web site.

The SPSS Statistical Procedures Companion, by Marija Norusis, has been published by Prentice Hall. A new version of this book, updated for SPSS 13.0, is planned

.TheSPSS Advanced Statistical Procedures Companion, also based on SPSS

13.0, is forthcoming. The SPSS Guide to Data Analysis forSPSS13.0isalsoin development. Announcements of publications available exclusively through Prentice Hall wi your home country, and then click

ll be available on the SPSS Web site at http://www.spss.com/estore (select

Books).

Tel l Us

Your co

Your Thoughts

mments are important. Please let us know about your experiences with SPSS products. We especially like to hear about new and interesting applications using the SPSS system. Please send e-mail to suggest@spss.com or write to SPSS Inc.,

Page 5

Attn.: Director of Product Planning, 233 South Wacker Drive, 11th Floor, Chicago, IL 60606-6412.

About This Manual

This manual documents the graphical user interface for the procedures included in the Complex Samples add-on module. Illustrations of dialog boxes are taken from SPSS for Win

dows. Dialog boxes in other operating systems are similar. Detailed information about the command syntax for features in this module is provided in the SPSS Command Syntax Reference, available from the Help menu.

Contacting SPSS

If you would like to be on our mailing list, contact one of our offices, listed on our Web site at http://www.spss.com/worldwide.

Page 6

Page 7

Contents

Part I: User

's Guide

1 Introduction to SPSS Complex Samples

Procedures 1

PropertiesofComplexSamples .................................. 1

UsageofComplexSamplesProcedures............................ 2

PlanFiles................................................ 3

FurtherReadings ............................................. 3

2 Sampling

CreatingaNewSamplePlan .................................... 6

SamplingWizard:DesignVariables ............................... 7

TreeControlsforNavigatingtheSamplingWizard................. 8

Sampling

Define Un

SamplingWizard:OutputVariables............................... 13

SamplingWizard:PlanSummary ................................ 15

SamplingWizard:DrawSampleSelectionOptions................... 16

SamplingWizard:DrawSampleOutputFiles........................ 17

SamplingWizard:Finish....................................... 18

ModifyinganExistingSamplePlan............................... 19

SamplingWizard:PlanSummary ................................ 20

from a Complex Design 5

Wizard:SamplingMethod............................... 9

Wizard:SampleSize.................................. 11

equalSizes...................................... 12

vii

Page 8

RunninganExistingSamplePlan ................................ 21

CSPLAN and CSSELECT Commands Additional Features . . . . . . . . . . . . . . . 21

3 Preparing a

CreatingaNewAnalysisPlan................................... 24

AnalysisPreparationWizard:DesignVariables...................... 25

TreeControlsforNavigatingtheAnalysisWizard................. 26

Analysis Pr

Define Uneq

AnalysisPreparationWizard:StageSummary ...................... 30

AnalysisPreparationWizard:Finish.............................. 31

ModifyinganExistingAnalysisPlan.............................. 32

AnalysisPreparationWizard:PlanSummary ....................... 33

ComplexSampleforAnalysis 23

eparationWizard:EstimationMethod.................... 27

eparationWizard:Size ............................... 28

ualSizes...................................... 29

4 Complex Samples Plan 35

5 Complex Samples Frequencies 37

ComplexSamplesFrequenciesStatistics.......................... 39

ComplexSamplesMissingValues................................ 40

ComplexSamplesOptions ..................................... 41

viii

Page 9

6 Complex Samp

ComplexSamplesDescriptivesStatistics.......................... 45

ComplexSamplesDescriptivesMissingValues...................... 46

ComplexSamplesOptions ..................................... 46

les Descriptives 43

7 Complex Sam

ComplexSamplesCrosstabsStatistics............................ 51

ComplexSamplesMissingValues................................ 53

ComplexSamplesOptions ..................................... 53

8 Complex Sa

ComplexSamplesRatiosStatistics............................... 57

ComplexSamplesRatiosMissingValues .......................... 58

ComplexSamplesOptions ..................................... 58

9 Complex S

Complex Sam ples General Linear Model Statistics . . . . . . . . . . . . . . . . . . . 65

ComplexSamplesHypothesisTests .............................. 66

Complex Sam ples General Linear Model Estimated Means . . . . . . . . . . . . . 68

ComplexSamplesGeneralLinearModelSave ...................... 69

Complex Sam ples General Linear Model Options . . . . . . . . . . . . . . . . . . . . 70

CSGLMCommandAdditionalFeatures............................ 71

mples Ratios 55

amples General Linear Model 61

ples Crosstabs 49

Page 10

10 Complex Samp

Complex Sam ples Logistic Regression Re ference Category . . . . . . . . . . . . 75

ComplexSamplesLogisticRegressionModel....................... 76

ComplexSamplesLogisticRegressionStatistics..................... 78

ComplexSamplesHypothesisTests .............................. 80

ComplexSamplesLogisticRegressionOddsRatios................... 81

ComplexSamplesLogisticRegressionSave........................ 83

ComplexSamplesLogisticRegressionOptions...................... 84

CSLOGISTICCommandAdditionalFeatures ........................ 85

les Logistic Regression 73

Part II: Exa

mples

11 Complex Samples Sampling Wizard 89

Obtaining

ObtainingaSamplefromaPartialSamplingFrame.................. 103

RelatedProcedures......................................... 121

aSamplefromaFullSamplingFrame..................... 89

Using the W

PlanSummary........................................... 99

SamplingSummary...................................... 100

Sample Res

UsingtheWizardtoSamplefromtheFirstPartialFrame .......... 103

Sample Resu

UsingtheWizardtoSamplefromtheSecondPartialFrame........ 114

SampleResults......................................... 120

izard ........................................ 89

ults......................................... 102

lts......................................... 113

Page 11

12 Complex Samp

les Analysis Preparation

Wizard 123

Using the Co

PublicData................................................ 123

UsingtheWizard........................................ 123

Summary.............................................. 126

Preparing for Analysis When Sampling Weights Are Not in the Data File. . 126

Computing Inclu s ion Probabilities and Sampling Weights . . . . . . . . . 127

Using the Wi

Summary.............................................. 137

Related Proc

mplex Samples Analysis Preparation Wizard to Ready NHIS

zard........................................ 130

edures ......................................... 137

13 Complex Samples Frequencies 139

Using Comp

Usage.................................................... 139

RelatedProcedures......................................... 144

lex S amp les Frequencies to Analyze N utritional Supplemen t

RunningtheAnalysis..................................... 139

Frequency

FrequencybySubpopulation............................... 143

Summary.............................................. 144

Table ........................................ 142

14 Complex Sa

Using Complex Samples Descriptives to Analyze Activity Levels . . . . . . . . 145

RunningtheAnalysis..................................... 145

UnivariateStatistics...................................... 148

mples Descriptives 145

Page 12

UnivariateStatisticsbySubpopulation........................ 149

Summary.............................................. 150

RelatedProcedures......................................... 150

15 Complex Sam

Using Complex Samples Crossta bs to Measure the Relative Risk of an

Event .................................................... 151

Running th

Crosstabulation......................................... 155

RiskEstimate .......................................... 155

Risk Estima

Summary.............................................. 157

Related Pro

cedures......................................... 158

ples Crosstabs 151

eAnalysis..................................... 151

tebySubpopulation............................. 157

16 Complex Samples Ratios 159

Using Comp

Running th

Ratios ................................................ 162

PivotedRatiosTable ..................................... 163

Summary.............................................. 16

RelatedProcedures......................................... 165

lex Samples Ratios to Aid Property Value Assessment . . . . . . 159

eAnalysis..................................... 159

17 Complex Samples General Linear Model 167

Using Complex Samples General Linear Model to Fit a Two-Factor ANOVA 167

RunningtheAnalysis..................................... 167

Model Summ

ary ........................................ 172

xii

Page 13

TestsofModelEffects.................................... 173

ParameterEstimates..................................... 174

Estimated Ma

Summary ............................................. 178

Related Proc

rginalMeans ................................ 175

edures ......................................... 179

18 Complex Samples Logistic Regression 181

Using Compl

Running the

PseudoR-Squares....................................... 187

Classification........................................... 187

Test s of Mod

ParameterEstimates..................................... 189

OddsRatios............................................ 190

Summary.............................................. 192

RelatedProcedures......................................... 192

Bibliograph

ex Samples Logistic Regression to Assess Credit Risk . . . . . . 181

Analysis..................................... 181

elEffects.................................... 188

y193

Index 195

xiii

Page 14

Page 15

Part 1: User's Guide

Page 16

Page 17

Chapter

Introductio

ntoSPSSComplex

Samples Procedures

An inherent assumption of analytical procedures in traditional software packages isthattheobservationsinadatafilerepresentasimplerandomsamplefromthe population of interest. This assumption is untenable for an increasing number of companies and researchers who find it both cost-effective and convenient to obtain samples in a more structured way.

The SPSS Complex Samples option allows you to select a sample according to a complex design and incorporate the design specifications into the data analysis, thus ensuring that your results are valid.

Properties of Complex Samples

A complex sample can differ from a simple random sample in many ways. In a simple random sample, individual sampling units are selected at random with e qual probability and without replacement (WOR) directly from the entire population. By contrast, a given complex sample can have some or all of the following features:

Stratification. Stratified sampling involves selecting samples independently within

non-overlapping subgroups of the population, or strata. For example, strata may be socioeconomic groups, job categories, age groups, or ethnic groups. With stratification, you can ensure adequate sample sizes for subgroups of interest, improve the precision of overall estimates, and use different sampling methods from stratum to stratum.

Clustering. Cluster sampling involves the selection of groups of sampling units, or

clusters. For example, clusters may be schools, hospitals, or geographical areas, and sampling units may be students, patients, or citizens. Clustering is common in multistage designs and area (geographic) samples.

Page 18

Chapter 1

Multiple stages. In multistage sampling, you select a first-stage sample based on

clusters. Then you create a second-stage sample by drawing subsamples from the selected cl

usters. If the second-stage sample is based on subclusters, you can then add a third stage to the sample. For example, in the first stage of a survey, a sample of cities could be drawn. Then, from the selected cities, households could be sampled. Finally, f

rom the selected households, individuals could be polled. The Sampling and

Analysis Preparation wizards allow you to specify three stages in a design.

Nonrandom sampling. When selection at random is difficult to obtain, units can be

sampled sy

Unequal selection probabilities. When sampling clusters that contain unequal numbers

stematically (at a fixed interval) or sequentially.

of units, you can use probability-proportional-to-size (PPS) sampling to make a cluster’

s selection probability equal to the proportion of units it contains. PPS

sampling can also use more general weighting schemes to select units.

Unrestricted sampling. Unrestricted sampling selects units with replacement (WR).

Thus, an i

Sampling weights. Sampling weights are automatically computed while drawing a

ndividual unit can be selected for the sample more than once.

complex sample and ideally correspond to the “frequency” that each sampling unit represe

nts in the target population. Therefore, the sum of the weights over the sample should estimate the population size. Complex Samples analysis procedures require sampling weights in order to properly analyze a complex sample. Note that these weights

should be used entirely within the Complex Samples option and should not be used with other analytical procedures via the Weight Cases procedure, which treats weightsascasereplications.

Usage of Complex Samples Procedures

Your usage of Complex Samples procedures depends on your particular needs. The primary types of users are those who:

 Plan and carry out surveys according to complex designs, possibly analyzing the

sample later. The primary tool for surveyors is the Sampling Wizard.

 Analyze sample data files previously obtained according to complex designs.

e using the Complex Samples analysis procedures, you may need to use the

Befor Analysis Preparation Wizard.

Page 19

Plan Files

Introduction t

Regardlessofwhichtypeofuseryouare,youneedtosupplydesigninformationto Complex Samples procedures. This information is stored in a plan file for easy reuse.

A plan file contains complex sample specifications. There are two types of plan files:

Sampling plan. The specifications given in the Sampling Wizard define a sample

design that specifications. The sampling plan file also contains a default analysis plan that uses estimation methods suitable for the specified sample design.

Analysis p

analysis procedures to properly compute variance estimates for a complex sample. The plan includes the sample structure, estimation methods for each stage, and reference Wizard allows you to create and edit analysis plans.

There are several advantages to saving your specifications in a plan file, including:

 A surveyor can specify the first stage of a multistage sampling plan and draw

first-st and then modify the sampling plan to include the second stage.

 Ananalystwhodoesn’thaveaccesstothesamplingplanfilecanspecifyan

analysis plan and refer to that plan from each Complex Samples analysis procedu

 Adesig

which simplifies the instructions for analysts and avoids the need for each analyst to specify his or her own analysis plans.

is used to draw a complex sample. The sampling plan file contains those

lan.

This plan file contains information needed by Complex Samples

s to required variables, such as sample weights. The Analysis Preparation

age units now, collect information on sampling units for the second stage,

re.

ner of large-scale public use samples can publish the sampling plan file,

o SPSS Complex Samples Procedures

Further

Readings

For more Cochran, W. G. 1977. Sampling Techniques. New York: John Wiley and Sons. Kish,L.1965.Survey Sampling. New York: John Wiley and Sons. Kish, L.

information on sampling techniques, see the following texts:

1987. Statistical Design for Research. New York: John Wiley and Sons.

Page 20

Chapter 1

Murthy, M. N. 1967. Sampling Theory and Methods. Calcutta, India: Statistical Publishing Society.

Särndal, C.

, B. Swensson, and J. Wretman. 1992. Model Assisted Survey Sampling.

New York: Springer-Verlag.

Page 21

Chapter

Sampling fro

Figure 2-1

Sampling W iz ard, Welcome step

maComplexDesign

The Sampling Wizard guides you through the steps for creating, modifying, or

uting a sampling plan file. Before using the Wizard, you should have a

exec well-defined target population, a list of sampling units, and an appropriate sample design in mind.

Page 22

Chapter 2

Creating a Ne

E From the me

Analyze

Complex Samples

Select a Sample...

Select Design a sample and choose a plan filename to save the sample plan.

E Click Next to continue through the Wizard. E Optionally, i n the Define Variables step, you can define strata, clusters, and input

sample weights. After you define these, click Next.

E Optionally,in the Sampling Method step, you can choose a method for selecting items.

If you select Otherwise, click Next and then:

E In the Sample Size step, specify the number or proportion of units to sample.

You can now click

 Choose output variables to save.  Add a second or third stage to the design.  Set various selection options, including which stages to draw samples from, the

random number seed, and whether to treat user-missing values as valid values of design var

 Choose wh  Paste your

wSamplePlan

nus choose:

PPS Brewer or PPS Murthy, you can click Finish to draw the sample.

Finish to draw the sample. Optionally, in further steps, you can:

iables.

eretosaveoutputdata. selections as command syntax.

Page 23

Sampling W iz

Figure 2-2

Sampling Wizard, Design Variables step

ard: Design Variables

Sampling from a

Complex Design

tep allows you to select stratification and clustering variablesand to define input

This s sample weights. You can also specify a label for the stage.

Stratify By. The cross-classification of stratification variables defines distinct

pulations, or strata. Separate samples are obtained for each stratum. To improve

subpo the precision of your estimates, units within strata should be as homogeneous as possible for the characteristics of interest.

ters.

Clus

are useful when directly sampling observational units from the population is expensive or impossible; instead, you can sample clusters from the population and then clusters can introduce correlations among sampling units, resulting in a loss of

Cluster variables define groups of observational units, or clusters. Clusters

sample observational units from the selected clusters. However, the use of

Page 24

Chapter 2

precision. To minimize this effect, units within clusters should be as heterogeneous as possible for the characteristics of interest. You must define at least one cluster variable in

order to plan a multistage design. Clusters are also necessary in the use of several different sampling methods. For more information, see “Sampling Wizard: Sampling Method” on p. 9.

Input Sampl

e Weight.

If the current sample design is part of a larger sample design, you may have sample weights from a previous stage of the larger design. You can specify a numeric variable containing these weights in the first stage of the current de

sign. Sample weights are computed automatically for subsequent stages

of the current design.

Stage Label. You can specify an optional string label for each stage. This is used in the

output to

help identify stagewise information.

Note: The source variable list has the same content across steps of the Wizard. In other words, variables removed from the source list in a particular step are removed from the list i

nallsteps.Variablesreturnedtothesource list appear in the list in all steps.

Tree Controls for Navigating the Sampling Wizard

On the left side of each step in the Sampling Wizard is an outline of all the steps. You can navi Steps are enabled as long as all previous steps are valid—that is, if each previous step has been given the minimum required specifications for that step. See the Help for individ

gatetheWizardbyclickingonthenameofanenabledstepintheoutline.

ual steps for more information on why a given step may be invalid.

Page 25

Sampling W iz

Figure 2-3

Sampling W izard, Method step

ard: Sampling Method

Sampling from a

Complex Design

tep allows you to specify how to select cases from the working data file.

This s

Method. Controls in this group are used to choose a selection method. Some sampling

types allow you to choose whether to sample with replacement (WR) or without

cement (WOR). See the type descriptions for more information. Note that some

repla probability-proportional-to-size (PPS) types are available only when clusters have been defined and that all PPS types are available only in the first stage of a design.

over, WR methods are available only in the last stage of a design.

 Sim

ple Random Sampling.

selected with or without replacement.

Units are selected with equal probability. They can be

Page 26

Chapter 2



Simple Systematic. Units are selected at a fixed interval throughout the sampling

frame (or st

rata,iftheyhavebeenspecified)and extracted without replacement.

A randomly selected unit w ithin the first interval is chosen as the starting point.

 Simple Sequential. Units are selected sequentially with equal probability and

without replacement.

 PPS. This is a first-stage method that selects units at random with probability

proportion

al to size. Any units can be selected with replacement; only clusters

can be sampled without replacement.

 PPS Systematic. This is a first-stage method that systematically selects units with

probability proportional to size. They are selected without replacement.

 PPS Sequential. This is a first-stage method that sequentially selects units with

probabili

 PPS Brewe

ty proportional to cluster size and without replacement.

This is a first-stage method that selects two clusters from each stratum with probability proportional to cluster size and without replacement. A cluster variable must be specified to use this method.

 PPS Murthy. This is a first-stage method that selects two clusters from each

stratum wi

th probability proportional to cluster size and without replacement. A

cluster variable must be specified to use this method.

 PPS Sampford. This is a first-stage method that selects more than two clusters

from each stratum with probability proportional to cluster size and without replacem

ent. It is an extension of Brewer’s method. A cluster variable must be

specified to use this method.

 Use WR estimation for analysis. By default, an estimation method is specified in

the plan file that is consistent with the selected sampling method. This allows you to use wi

th-replacement estimation even if the sampling method implies WOR

estimation. This option is available only in stage 1.

Measure

of Size (MOS).

If a PPS method is selected, you must specify a measure of size that defines the size of each unit. These sizes can be explicitly defined in a variable or they can be computed from the data. Optionally, you can set lower and upper b

oundsontheMOS,overridinganyvaluesfoundintheMOSvariableor

computed from the data. These options are available only in stage 1.

Page 27

Sampling Wiz

Figure 2-4

Sampling Wizard, Sample Size step

ard: Sample Size

Sampling from a

Complex Design

tep allows you to specify the number or proportion of units to sample within

This s the current stage. The sample size can be fixed or it can vary across strata. For the purpose of specifying sample size, clusters chosen in previous stages can be used to

ne strata.

defi

Units. You can specify an exact sample size or a proportion of units to sample.

 Value. A single value is applied to all strata. If Counts is selected as the unit

ic, you should enter a positive integer. If

metr

Proportions is selected, you should

enter a non-negative value. Unless sampling with replacement, proportion values should also be no greater than 1.

Page 28

Chapter 2



Unequal values for strata. Allows you to enter size values on a per-stratum basis

via the Defi

 Read values

ne Unequal Sizes dialog box.

from variable.

Allows you to select a numeric variable that contains

size values for strata.

is selected, you have the option to set lower and upper bounds on

Proporti

ons

the number of units sampled.

Define Uneq

Figure 2-5

Define Unequal Sizes dialog box

The Define Unequal Sizes dialog box allows you to enter sizes on a per-stratum basis.

Size Specifications grid. The grid displays the cross-classifications of up to five strata

or cluster variables—one stratum/cluster combination per row. Eligible grid variables include all stratification variables from the current and previous stages and all cluster variables from previous stages. Variables can be reordered within the grid or moved to the Exclude list. Enter sizes in the rightmost column. Click to toggle the display of value labels and data values for stratification and cluster variables in the grid cells. Cells that contain unlabeled values always show values.

ual Sizes

Labels or Values

Page 29

Click Refresh Strata to repopulate the grid with each combination of labeled data values for variables in the grid.

Exclude. To

more variables to the Exclude list. These variables are not used to define sample sizes.

Sampling Wi

Figure 2-6

Sampling Wizard, Output Variables step

Sampling from a

specify sizes for a subset of stratum/cluster combinations, move one or

Complex Design

zard: Output Variables

tep allows you to choose variables to save when the sample is drawn.

This s

Population size. The estimated number of units in the population for a given stage.

The root name for the saved variable is PopulationSize_.

e proportio n.

Sampl

variable is SamplingRate_.

The sampling rate at a given stage. The root name for the saved

Page 30

Chapter 2

Sample size. The number of units drawn at a given stage. The root name for t he

saved variable is SampleSize_.

ht.

Sample weig

The inverse of the inclusion probabilities. The root name for the

saved variable is SampleWeight_.

Some stagewise variables are generated automatically. These include:

Inclusion probabilities. The proportion of units drawn at a given stage. The root name

for the save

Cumulative weight. The cumulative sample weight over stages previous to

dvariableisInclusionProbability_.

and including the current one. The root name for the saved variable is

SampleWei

Index. Identifies units selected multiple times within a given stage. The root name for

ghtCumulative_.

the saved variable is Index_. Note:Save

d variable root names include an integer suffix that reflects the stage

number—for example, PopulationSize_1_ for the saved population size for stage 1.

Page 31

Sampling Wiz

Figure 2-7

Sampling Wizard, Plan Summary step

ard: Plan Summary

Sampling from a

Complex Design

s the last step within each stage, providing a summary of the sample design

This i specifications through the current stage. From here, you can either proceed to the next stage (creating it, if necessary) or set options for drawing the sample.

Page 32

Chapter 2

Sampling Wiz

Figure 2-8

Sampling Wizard, Draw Sample, Selection Options step

ard: Draw Sample Selection Options

tep allows you to choose whether to draw a sample. You can also control other

This s sampling options, such as the random seed and missing-value handling.

Draw sample. In addition to choosing whether to draw a sample, you can also choose

cute part of the sampling design. Stages must be drawn in order—that is,

to exe stage 2 cannot be drawn unless stage 1 is also drawn. When editing or executing a plan, you cannot resample locked stages.

This allows you to choose a seed value for random number generation.

Seed

Include user-missing values. This determines whether user-missing values are valid. If

so, user-missing values are treated as a separate category.

already sorted.

Data

variables, this option allows you to speed the selection process.

If your sample frame is presorted by the values of the stratification

Page 33

Sampling Wiz

Figure 2-9

Sampling Wizard, Draw Sample, Output Files step

Sampling from a

ard: Draw Sample Output Files

Complex Design

tep allows you to choose where to direct sampled cases, weight variables, joint

This s probabilities, and case selection rules.

Sample da ta. These options let you determine where sample output is written. It can

ed to the working data file or saved to an external file. If an external file is

be add specified, the sampling output variables and variables in the working data file for the selected cases are saved to the file.

t probabilities.

Join

written. Joint probabilities are produced if the PPS WOR, PPS Brewer, PPS Sampford, or PPS Murthy method is selected and WR estimation is not specified.

These options let you determine where joint probabilities are

Page 34

Chapter 2

Case selection rules. If you are constructing your sample one stage at a time, you may

want to save the case selection rules to a text file. They are useful for constructing the subframe fo

r subsequent stages.

Sampling Wizard: Finish

Figure 2-10

Sampling Wizard, Finish step

This is the final step. You can save the plan file and draw the sample now or paste

selections into a syntax window.

your

When making changes to stages in the existing plan file, you can save the edited

plan to a new file or overwrite the existing file. When adding stages without making

ges to existing stages, the Wizard automatically overwrites the existing plan file.

chan Ifyouwanttosavetheplantoanewfile,select

Wizard into a syntax window

and change the filename in the syntax commands.

Paste the syntax generated by the

Page 35

Modifying an

E From the me

Analyze

Complex Samples

Select a Sample...

Select Edit a sample design and choose a plan file to edit.

E Click Next to continue through the Wizard. E Review the sampling plan in the Plan Summary step, and then click Next.

Subsequent s steps for more information.

E Navigate to the Finish step, and specify a new name for the edited plan file or choose

to overwrite the existing plan file.

Optionally, you can:

 Specify stages that have already been sampled.  Remove stages from the plan.

Sampling from a

Complex Design

Existing Sample Plan

nus choose:

teps are largely the same as for a new design. See the Help for individual

Page 36

Chapter 2

Sampling Wiz

Figure 2-11

Sampling Wizard, Plan Summary step

ard: Plan Summary

tep allows you to review the sampling plan and indicate stages that have already

This s been sampled. If editing a plan, you can also remove stages from the plan.

Previously sampled stages. If an extended sampling frame is not available, you will

o execute a multistage sampling design one stage at a time. Select which stages

have t have already been sampled from the drop-down list. Any stages that have been executed are locked; they are not available in the Draw Sample Selection Options

, and they cannot be altered when editing a plan.

step

Remove stages. You can remove stages 2 and 3 from a multistage design.

Page 37

Complex Design

RunninganEx

E From the me

Analyze

Complex Samples

Select a Sample...

Select Draw a sample and choose a plan file to run.

E Click Next to continue through the Wizard. E Review the sampling plan in the Plan Summary step, and then click Next. E The individual steps containing stage information are skipped when executing a

Sampling from a

isting Sample Plan

nus choose:

sample plan. You can now go on to the Finish step at any time.

Optionally, you can:

 Specify stages that have already been sampled.

CSPLAN and CSSELECT Commands Additional Features

The SPSS command language also allows you to:

 Specify custom names for output variables.  Control the output in the Viewer. For example, you can suppress the stagewise

summary of t

he plan that is displayed if a sample is designed or modified,

suppress the summary of the distribution of sampled cases by strata that is shown if the sample design is executed, and request a case processing summary.

 Choose a subset of variables in the working data file to write to an external

sample fil

See the SPSS Command Syntax Reference for complete syntax information.

Page 38

Page 39

Chapter

Preparing a C for Analysis

Figure 3-1

Analysis Preparation Wizard, Welcome step

omplex Sample

Page 40

Chapter 3

The Analysis Preparation Wizard guides you through the steps for creating o r modifying an analysis plan for use with the various Complex Samples analysis procedures

. Before using the Wizard, you should have a sample drawn according to a

complex design.

Creating a new plan is most useful when you do not have access to the sampling

plan file us

ed to draw the sample (recall that the sampling plan contains a default analysis plan). If you do have access to the sampling plan file used to draw the sample, you can use the default analysis plan contained in the sampling plan file or override t

he default analysis specifications and save your changes to a new file.

Creating a New Analysis Plan

E From the menus choose:

Analyze

Complex Sa

Prepare fo

mples

r Analysis...

Select Cr

eateaplanfile

, and choose a plan filename to which you will save the

analysis plan.

E Click Next to continue through the Wizard. E Specify the variable containing sample weights in the Design Variables step,

optionally defining strata and clusters.

You can now click

 Select the method for estimating standard errors in the Estimation Method step.  Specify the number of units sampled or the inclusion probability per unit in

Finish to save the plan. Optionally, in further steps you can:

the Size step.

 Add a second or third stage to the design.  Paste your selections as command syntax.

Page 41

Analysis Pre

Figure 3-2

Analysis Preparation Wizard, Design Variables step (stage 1)

Preparing a Com

paration Wizard: Design Variables

plex Sample for Ana lysis

tep allows you to identify the stratification and clustering variables and define

This s sample weights. You can also provide a label for the stage.

Strata. The cross-classification of stratification variables defines distinct

pulations, or strata. Your total sample represents the combination of

subpo independent samples from each stratum.

Clusters. Cluster variables define groups of observational units, or clusters. Samples

ninmultiplestagesselectclustersintheearlier stages and then subsample units

draw from the selected clusters. When analyzing a data file obtained by sampling clusters with replacement, you should include the duplication index as a cluster variable.

le Weights.

Samp

are computed automatically for subsequent stages of the current design.

You must provide sample weights in the first stage. Sample weights

Page 42

Chapter 3

Stage Label. You can specify an optional string label for each stage. This is used in the

output to help identify stagewise information. Note: The so

urce variable list has the same contents across steps of the Wizard. In other words, variables removed from the source list in a particular step are removed from the list in all steps. Variables returned to the source list show up in all steps.

Tree Controls for Navigating the Analysis Wizard

At the left side of each step of the Analysis Wizard is an outline of all the steps. You can navigate the Wizard by clicking on the name of an enabled step in the outline. Steps are e previous step has been given the minimum required specifications for that step. For more information on why a given step may be invalid, see the Help for individual steps.

nabled as long as all previous steps are valid—that is, as long as each

Page 43

Analysis Pre

Figure 3-3

Analysis Preparation Wizard, Estimation Metho d step (stage 1)

Preparing a Com

paration Wizard: Estimation Method

plex Sample for Ana lysis

tep allows you to specify an estimation method for the stage.

This s

WR (sampling with replacement). WR estimation does not include a correction for

sampling from a finite population, since it assumes that the sample was taken from

inite population. When the population for the stage is much larger than the

an inf sample, this is a reasonable assumption. WR estimation can be specified only in the final stage of a design; the Wizard will not allow you to add another stage if you

ct WR estimation.

sele

Equal WO R (equal probability sampling without replacement). Equal WOR estimation

includes the finite population correction and assumes that units are sampled with

l probability. Equal WOR can be specified in any stage of a design.

equa

Page 44

Chapter 3

Unequal WOR (unequal probability sampling without replacement). In addition to using

the finite population correction, Unequal WOR accounts for sampling units (usually clusters) s only in the first stage.

elected with unequal probability. This estimation method is available

Analysis Pr

Figure 3-4

Analysis Preparation Wizard, Size step (stage 1)

eparation Wizard: Size

step is used to specify inclusion probabilities or population sizes for the current

This stage. Sizes can be fixed or can vary across strata. For the purpose of specifying sizes, clusters specified in previous stages can be used to define strata.

You can specify exact population sizes or the probabilities with which units

Unit

were sampled.

Page 45



Value. A single value is applied to all strata. If Population Sizes is selected as the

unit metric selected, you should enter a value between 0 and 1, inclusive.

 Unequal values for strata. Allows you to enter size values on a per-stratum basis

via the Define Unequal Sizes dialog box.

 Read values from variable. Allows you to select a numeric variable that contains

size values

Define Unequal Sizes

Figure 3-5

Define Unequal Sizes dialog box

Preparing a Com

, you should enter a non-negative integer. If

for strata.

plex Sample for Ana lysis

Inclusion Probabilities is

The Define Unequal Sizes dialog box allows you to enter sizes on a per-stratum basis.

Size Specifications grid. The grid displays the cross-classifications of up to five strata

Labels or Values

to toggle the display of value labels and data values for stratification and cluster variables in the grid cells. Cells that contain unlabeled values always show values.

Page 46

Chapter 3

Click Refresh Strata to repopulate the grid with each combination of labeled data values for variables in the grid.

Exclude. To

more variables to the Exclude list. These variables are not used to define sample sizes.

Analysis Pr

Figure 3-6

Analysis Preparation Wizard, Plan Summar y step (stage 1)

specify sizes for a subset of stratum/cluster combinations, move one or

eparation Wizard: Stage Summary

s the last step within each stage, providing a summary of the analysis design

This i specifications through the current stage. From here, you can either proceed to the next stage (creating it if necessary) or save the analysis specifications.

If you cannot add another stage, it is likely because:

 No cluster variable was specified in the Design Variables step.

Page 47



You selected WR estimation in the Estimation Method step.

 This is the third stage of the analysis, and the Wizard supports a maximum of

three stages

Analysis Preparation Wizard: Finish

Figure 3-7

Analysis Preparation Wizard, Finish step

Preparing a Com

plex Sample for Ana lysis

This is the final step. You can save the plan file now or paste your selections to

ax window.

asynt

When making changes to stages in the existing plan file, you can save the edited

plan to a new file or overwrite the existing file. When adding stages without making

ges to existing stages, the Wizard automatically overwrites the existing plan file.

chan If you want to save the plan to a new file, choose to

Wizard into a syntax window

and change the filename in the syntax commands.

Paste the syntax g enerated by the

Page 48

Chapter 3

Modifying an

E From the me

Analyze

Complex Samples

Prepare for Analysis...

Select Edit a plan file, and choose a plan filename to which you will save the analysis

plan.

E Click Next to continue through the Wizard. E Review the analysis plan in the Plan Summary step, and then click Next.

Subsequent steps are largely the same as for a new design. For more information, see the Help for individual steps.

E Navigate to the Finish step, and specify a new name for the edited plan file, or choose

to overwrit

Optionally

 Remove st

Existing Analysis Plan

nus choose:

e the existing plan file.

, you can:

ages from the plan.

Page 49

Analysis Pre

Figure 3-8

Analysis Preparation Wizard, Plan Summary step

Preparing a Com

paration Wizard: Plan Summary

plex Sample for Ana lysis

tep allows you to review the analysis plan and remove stages from the plan.

This s

Remove Stages. You can remove stages 2 and 3 from a multistage design. Since a plan

must have at least one stage, you can edit but not remove stage 1 from the design.

Page 50

Page 51

Chapter

Complex Samp

Complex Samples analysis procedures require analysis specifications from an analysis or sample plan file in order to provide valid results.

Figure 4-1

Complex Samples Plan dialog box

les Plan

Plan. Specify the path of an analysis or sample plan file. Joint Probabilities. In order to use Unequal WOR estimation for clusters drawn

using a PPS WOR method, you need to specify a separate file containing the joint probabilities. This file is created by the Sampling Wizard during sampling.

Page 52

Page 53

Chapter

Complex Samp

The Complex Samples Frequencies procedure produces frequency tables for selected variables a nd displays univariate statistics. Optionally, you can request statistics by subgroups, defined by one or more categorical variables.

Example. Using the Complex Samples Frequencies procedure, you can obtain

univariate tabular statistics for vitamin usage among U.S. citizens, based on the results of the National Health Interview Survey (NHIS) and with an appropriate analysis plan for this public use data.

Statistics. The procedure produces estimates of cell population sizes and table

percentages, plus standard errors, confidence intervals, coefficients of variation, design effects, square roots of design effects, cumulative values, and unweighted counts for each estimate. Additionally, chi-square and likelihood ratio statistics are computed for the test of equal cell proportions.

Data. Variables for which frequency tables are produced should be categorical.

Subpopulation variables can be string or numeric, but should be categorical.

Assumptions. The cases in the data file represent a sample from a complex design

that should be analyzed according to the specifications in the file selected in the Complex Samples Plan dialog box.

les Frequencies

Obtaining Complex Samples Frequencies

E From the menus choose:

Analyze

Complex Samples

Frequencies...

Select a plan file and optionally select a custom joint probabilities file.

Page 54

Chapter 5

Click Continue.

Figure 5-1

Frequencies

E Select at least one frequency variable.

dialog box

Optionally, you can:

 Specify variables to define subpopulations. Statistics are computed separately for

each subpopulation.

Page 55

Complex Samp

Figure 5-2

Frequencies Statistics dialog box

Cells. This group allows you to request estimates of the cell population sizes and

table percentages.

Statistics. This group produces statistics associated with the population size or table

percentage.

 Standard error. The standard error of the estimate.  Confidence interval. A confidence interval for the estimate, using the specified

level.

 Coefficient of variation. The ratio of the standard error of the estimate to the

estimate.

 Unweighted count. The number of units used to compute the estimate.  Design effect. The ratio of the variance of the estimate to the variance obtained

by assuming that the sample is a simple random sample. This is a measure of the effect of specifying a complex design, where values further from 1 indicate greater effects.

 Square root of design effect. This is a measure of the effectof specifying a complex

design, where values further from 1 indicate greater effects.

 Cumulative values. The cumulative estimate through each value of the variable.

les Frequencies Statistics

Complex Sample

s Frequencies

Page 56

Chapter 5

Test of equal cell proportions. This produces chi-square and likelihood-ratio tests of

the hypothesis that the categories of a variable have equal frequencies. Separate tests are pe

rformed for each variable.

Complex Samples Missing Values

Figure 5-3

Missing Values dialog box

Tables.

 Use all

This group determines which cases are used in the analysis.

available data.

Missing values are determined on a table-by-table basis. Thus, the cases used to compute statistics may vary across frequency or crosstabulation tables.

 Use consistent case base. Missing values are determined across all variables.

he cases used to compute statistics are consistent across tables.

Thus, t

Categorical Desig n Variables. This group determines whether user-missing values are

valid or invalid.

Page 57

Complex Samp

Figure 5-4

Options dialog box

Subpopulation Display. You can choose to have subpopulations displayed in the same

table or in separate tables.

les Options

Complex Sample

s Frequencies

Page 58

Page 59

Chapter

Complex Samp

The Complex Samples Descriptives procedure displays univariate summary statistics for several variables. Optionally, you can request statistics by subgroups, defined by one or more categorical variables.

Example. Using the Complex Samples Descriptives procedure, you can obtain

univariate descriptive statistics for the activity levels of U.S. citizens based on the results of the National Health Interview Survey (NHIS) and with an appropriate analysis plan for this public use data.

Statistics. The procedure produces means and sums, plus t tests, standard errors,

confidence intervals, coefficients of variation, unweighted counts, population sizes, design effects, and square roots of design effects for each estimate.

Data. Measures should be scale variables. Subpopulation variables can be string

or numeric but should be categorical.

Assumptions. The cases in the data file represent a sample from a complex design

that should be analyzed according to the specifications in the file selected in the Complex Samples Plan dialog box.

les Descriptives

Obtaining Complex Samples Descriptives

E From the menus choose:

Analyze

Complex Samples

Descriptives...

Select a plan file, and optionally select a custom joint probabilities file.

E Click Continue.

Page 60

Chapter 6

Figure 6-1

Descriptives dialog box

E Select at least one measure variable.

Optionally, you can:

 Specify variables to define subpopulations. Statistics are computed separately for

each subpopulation.

Page 61

Complex Samp

Figure 6-2

Descriptives Statistics dialog box

Summaries. This group allows you to request estimates of the means and sums of the

measure a specified value.

variables. Additionally, you can request t tests of the estimates against

les Descriptives Statistics

Complex Sample

s Descriptives

Statistics. This group produces statistics associated with the mean or sum.

 Standard error. The standard error of the estimate.  Confidence interval. A confidence interval for the estimate, using the specified

level.

 Coeff

icient of variation.

The ratio of the standard error of the estimate to the

estimate.

 Unweighted count. The number of units used to compute the estimate.  Population size. The estimated number of units in the population.  Design effect. The ratio of the variance of the estimate to the variance obtained

by assuming that the sample is a simple random sample. This is a measure of

ect of specifying a complex design, where values further from 1 indicate

the eff greater effects.

 Square root of design effect. This is a measure of the effectof specifying a complex

design, where values further from 1 indicate greater effects.

Page 62

Chapter 6

Complex Samp

Figure 6-3

Descriptives Missing Values dialog box

Statistics for Measure Variables. This group determines which cases are used in the

analysis.

 Use all available data. Missing values are determined on a variable-by-variable

basis, thus the cases used to compute statistics may vary across measure variables.

 Ensure consistent case base. Missing values are determined across all variables,

thus the

les Descriptives Missing Values

cases used to compute statistics are consistent.

Categorical Design Variables. This group determines whether user-missing values are

valid or invalid.

Complex Samples Options

Figure 6-4

Options dialog box

Page 63

Complex Sample

s Descriptives

Subpopulation Display. You can choose to have subpopulations displayed in the same

table or in separate tables.

Page 64

Page 65

Chapter

Complex Samp

The Complex Samples Crosstabs procedure produces crosstabulation tables for pairs of selected variables and displays two-way statistics. Optionally, you can request statistics by subgroups, defined by one or more categorical variables.

Example. Using the Complex Samples Crosstabs procedure, you can obtain

cross-classification statistics for smoking frequency by vitamin usage of U.S. citizens, based on the results of the National Health Interview Survey (NHIS) and with an appropriate analysis plan for this public-use data.

Statistics. The procedure produces estimates of cell population sizes and row, column,

and table percentages, plus standard errors, confidence intervals, coefficients of variation, expected values, design effects, square roots of design effects, residuals, adjusted residuals, and unweighted counts for each estimate. The odds ratio, relative risk, and risk difference are computed for 2-by-2 tables. Additionally, Pearson and likelihood-ratio statistics are computed for the test of independence of the row and column variables.

Data. Row and column variables should be categorical. Subpopulation variables can

be string or numeric but should be categorical.

les Crosstabs

Assumptions. The cases in the data file represent a sample from a complex design

that should be analyzed according to the specifications in the file selected in the Complex Samples Plan dialog box.

Obtaining Complex Samples Crosstabs

E From the menus choose:

Analyze

Complex Samples

Crosstabs...

Page 66

Chapter 7

Select a plan file and, optionally, select a custom joint probabilities file.

E Click Continue.

Figure 7-1

Crosstabs di

E Select at least one row variable and one column variable.

alog box

Optionally, you can:

 Specify variables to define subpopulations. Statistics are computed separately for

each subpopulation.

Page 67

Complex Samp

Figure 7-2

Crosstabs Statistics dialog box

les Crosstabs Statistics

Complex Sample

s Crosstabs

Cells. This group allows you to request estimates of the cell population size and row,

column, and table percentages.

Statistics. This group produces statistics associated with the population size and row,

column, and table percentages.

 Standard error. The standard error of the estimate.  Confidence interval. A confidence interval for the estimate, using the specified

level.

 Coefficient of variation. The ratio of the standard error of the estimate to the

estimate.

 Expected values. The expected value of the estimate, under the hypothesis of

independence of the row and column variable.

 Unweighted count. The number of units used to compute the estimate.

Page 68

Chapter 7



Design effe ct. The ratio of the variance of the estimate to the variance obtained

by assuming

thatthesampleisasimplerandomsample.Thisisameasureof the effect of specifying a complex design, where values further from 1 indicate greater effects.

 Square root of design effect. This is a measure of the effectof specifying a complex

design, whe

 Residuals

re values further from 1 indicate greater effects.

The expected value is the number of cases that you would expect in the cell if there were no relationship between the two variables. A positive residual indicates that there are more cases in the cell than there would be if the row and column var

 Adjusted

iables were independent.

residuals.

The residual for a cell (observed minus expected value) dividedbyanestimateofitsstandarderror.Theresultingstandardizedresidualis expressed in standard deviation units above or below the mean.

Summaries for 2-by-2 Tables. This group produces statistics for tables in which the row

and column variable each have two categories. Each is a measure of the strength of the assoc

 Odds rat

iation between the presence of a factor and the occurrence of an event.

io.

The odds ratio can be used as an estimate of relative risk when the

occurrence of the factor is rare.

 Relative risk. The ratio of the risk of an event in the presence of the factor to the

risk of the event in the absence of the factor.

 Risk difference. Thedifferencebetweentheriskofaneventinthepresenceofthe

factor an

d the risk of the event in the absence of the factor.

Test of i

ndependence of rows and columns.

This produces chi-square and likelihood-ratio tests of the hypothesis that a row and column variable are independent. Separate tests are performed for each pair of variables.

Page 69

Complex Samp

Figure 7-3

Missing Values dialog box

Tables. This group determines which cases are used in the analysis.

 Use all available data. Missing values are determined on a table-by-table basis.

Thus, th crosstabulation tables.

 Use consistent case base. Missing values are determined across all variables.

Thus, the cases used to compute statistics are consistent across tables.

Complex Sample

les Missing Values

e cases used to compute statistics may vary across frequency or

s Crosstabs

Categorical Desig n Variables. This group determines whether user-missing values are

valid or

invalid.

Complex Samples Options

Figure 7-4

Options dialog box

Page 70

Chapter 7

Subpopulation Display. You can choose to have subpopulations displayed in the same

table or in separate tables.

Page 71

Chapter

Complex Samp

The Complex Samples Ratios procedure displays univariate summary statistics for ratios of variables. Optionally, you can request statistics by subgroups, defined by one or more categorical variables.

Example. Using the Complex Samples Ratios procedure, you can obtain descriptive

statistics for the ratio of current property value to last assessed value, based on the results of a statewide survey carried out according to a complex design and with an appropriate analysis plan for the data.

Statistics. The procedure produces ratio estimates, t tests, standard errors, confidence

intervals, coefficients of variation, unweighted counts, population sizes, design effects, and square roots of design effects.

Data. Numerators and denominators should be positive-valued scale variables.

Subpopulation variables can be string or numeric but should be categorical.

Assumptions. The cases in the data file represent a sample from a complex design

that should be analyzed according to the specifications in the file selected in the Complex Samples Plan dialog box.

les Ratios

Obtaining Complex Samples Ratios

E From the menus choose:

Analyze

Complex Samples

Ratios...

Select a plan file and, optionally, select a custom joint probabilities file.

E Click Continue.

Page 72

Chapter 8

Figure 8-1

Ratios dialog box

E Select at least one numerator variable and denominator variable.

Optionally, you can:

 Specify variables to define subgroups for which statistics are produced.

Page 73

Complex Samp

Figure 8-2

Ratios Statistics dialog box

ics.

Statist

 Standa

 Confide

level.

 Coefficient of variation. The ratio of the standard error of the estimate to the

estimate.

by assum the effect of specifying a complex design, where values further from 1 indicate greater effects.

 Square root of design effect. This is a measure of the effectof specifying a complex

design,

Complex Sample

les Ratios Statistics

This group produces statistics associated with the ratio estimate.

rd error.

nce interval.

The standard error of the estimate.

A confidence interval for the estimate, using the specified

ing that the sample is a simple random sample. This is a measure of

where values further from 1 indicate greater effects.

sRatios

You can request t tests of the estimates against a specified value.

t-test

Page 74

Chapter 8

Complex Samp

Figure 8-3

Ratios Missing Values dialog box

Ratios. This group determines which cases are used in the analysis.

 Use all available da ta. Missing values are determined on a ratio-by-ratio basis.

Thus, th pairs.

 Ensure consistent case base. Missing values are determined across all variables.

Thus, the cases used to compute statistics are consistent.

les Ratios Missing Values

e cases used to compute statistics may vary across numerator-denominator

Categorical Design Variables. This group determines whether user-missing values are

valid or invalid.

Complex Samples Options

Figure 8-4

Options dialog box

Page 75

Complex Sample

sRatios

Subpopulation Display. You can choose to have subpopulations displayed in the same

table or in separate tables.

Page 76

Page 77

Chapter

Complex Samp

les

General Linear Model

The Comple regression analysis, as well as analysis of variance and covariance, for samples drawn by complex sampling methods. Optionally, you can request analyses for a subpopul

Example. A grocery store chain surveyed a set of customers concerning their

purchasing habits, according to a complex design. Given the survey results and how much e frequency with which customers shop is related to the amount they spend in a month, controlling for the gender of the customer and incorporating the sampling design.

Statisti

t tests, design effects, and square roots of design effects for model parameters, as well as the correlations and covariances between parameter estimates. Measures of model fi also available. Additionally, you can request estimated marginal means for levels of model factors and factor interactions.

x Samples General Linear Model (CSGLM) procedure performs linear

ation.

ach customer spent in the previous month, the store wants to see if the

cs.

The procedure produces estimates, standard errors, confidence intervals,

t and descriptive statistics for the dependent and independent variables are

he dependent variable is quantitative. Factors are categorical. Covariates

Data. T

are quantitative variables that are related to the dependent variable. Subpopulation variables can be string or numeric but should be categorical.

tions.

Assump

that should be analyzed according to the specifications in the file selected in the Complex Samples Plan dialog box.

Thecasesinthedatafilerepresentasamplefromacomplexdesign

Page 78

Chapter 9

Obtaining a Co mplex Samples General Linear Model

From the menus choose:

Analyze

Complex Samples

General Linear Model...

Select a plan file and, optionally, select a custom joint probabilities file.

E Click Continue.

Figure 9-1

General Linear Model dialog box

E Select a dependent variable.

Page 79

Complex Sample

s General Linear Model

Optionally, you can:

 Select variables for Factors and Covariates, as appropriate for your data.  Specify a variable to define a subpopulation. The analysis is performed only for

the selected

Figure 9-2

Model dialog box

category of the subpopulation variable.

Specify Model Effects. By default, the procedure builds a main-effects model using the

factors and covariates specified in the main dialog box. Alternatively, you can build a custom model that includes interaction effects and nested terms.

Page 80

Chapter 9

Non-Nested Terms

For the selected factors and covariates:

Interaction. Creates the highest-level interaction term for all selected variables.

Main effects

All 2-way. Creates all possible two-way interactions of the selected variables. All 3-way. Creates all possible three-way interactions of the selected variables.

Creates a main-effects term for each variable selected.

All 4-way. Cr All 5-way. Creates all possible five-way interactions of the selected variables.

Nested Terms

You can buil

eates all possible four-way interactions of the selected variables.

d nested terms for your model in this procedure. Nested terms are useful for modeling the effect of a factor or covariate whose values do not interact with the levels of another factor. For example, a grocery store chain may follow the spending ha

bits of its customers at several store locations. Since each customer

frequents only one of these locations, the Customer effect can be said to be nested within the Store location effect.

Additiona

lly, you can include interaction effects or add multiple levels of nesting

to the nested term.

Limitations. Nested terms have the following restrictions:

 All factors within an interaction must be unique. Thus, if A is a factor, then

specifyin

 All facto

g A*A is invalid.

rs within a nested effect must be unique. Thus, if A is a factor, then

specifying A(A) is invalid.

 No effect can be nested within a covariate. Thus, if A is a factor and X is a

covariate, then specifying A(X) is invalid.

Intercept. The intercept is usually included in the model. If you can assume the

data pass through the origin, you can exclude the intercept. Even if you include the intercep

t in the model, you can choose to suppress statistics related to it.

Page 81

Complex Samp

Figure 9-3

General Linear Model Statistics dialog box

Model Parameters. This group allows you to control the display of statistics related to

the mod

 Estim

 Stand

 Confid

 t-test. Displays a t test of each coefficient estimate. The null hypothesis for each

 Covariances of parameter estimates. Displays an estimate of the covariance matrix

 Corre

el parameters.

ate.

ard error.

The confidence level for the interval is set in the Options dialog box.

test is that the value of the coefficient is 0.

for the

lations of parameter estimates.

for the model coefficients.

Complex Sample

les General Linear Model Statistics

Displays estimates of the coefficients.

Displays the standard error for each coefficient estimate.

ence interval.

model coefficients.

Displays a confidence interval for each coefficient estimate.

Displays an estimate of the correlation matrix

s General Linear Model

Page 82

Chapter 9



Design effe ct. The ratio of the variance of the estimate to the variance obtained

by assuming

thatthesampleisasimplerandomsample.Thisisameasureof the effect of specifying a complex design, where values further from 1 indicate greater effects.

 Square root of design effect. This is a measure of the effectof specifying a complex

design, whe

Model fit. D

re values further from 1 indicate greater effects.

isplays R

and root mean squared error statistics.

Population

means of dependent variable and covariates.

about the dependent variable, covariates, and factors.

Sample design information. Displays summary information about the sample,

including the unweighted count and the population size.

Complex Samples Hypothesis Tests

Figure 9-4

Hypothesis Tests d ialog box

Displays summary information

Page 83

Complex Sample

Test Statistic. This group allows you to select the type of statistic used for testing

s General Linear Model

hypotheses. You can choose between F, adjusted F, chi-square, and adjusted chi-square

Sampling De

grees of Freedom.

This group gives you control over the sampling design degrees of freedom used to compute p values for all test statistics. If based on the sampling design, the value is the difference between the number of primary sampling units and t

he number of strata in the first stage of sampling. Alternatively, you can set

a custom degrees of freedom by specifying a positive integer.

Adjustment for Multiple Comparisons. When performing many hypothesis tests, you

run the risk of increased overall Type I error (the probability of incorrectly rejecting a null hyp

othesis). This group allows you to choose the method of adjusting the

significance level.

 Least significant difference. This method does not control the overall probability

of reject

ing the hypotheses that some linear contrasts are different from the

null hypothesis values.

 Sequential Sidak. This is a sequentially step-down rejective Sidak procedure

that is much less conservative in terms of rejecting individual hypotheses but maintai

 Sequen

ns the same overall significance level.

tial Bonferroni.

This is a sequentially step-down rejective Bonferroni procedure that is much less conservative in terms of rejecting individual hypotheses but maintaining the same overall significance level.

 Sidak. This method provides tighter bounds than the Bonferroni approach.  Bonferroni. This method adjusts the observed significance level for the fact that

e contrasts are being tested.

multipl

Page 84

Chapter 9

Complex Samp

Figure 9-5

General Linear Model Esti mated Means dialog box

The Estimated Means dialog box allows you to display the model-estimated marginal means for levels of factors and factor interactions specified in the Model subdialog box. You can also request that the overall population mean be displayed.

les General Linear Model Estimated Means

Term . Estimated means are computed for the selected factors and factor interactions.

Contrast. The contrast determines how hypothesis tests are set up to compare the

estimated means.

 Simple. Compares the mean of each level to the mean of a specified level. This

type of contrast is useful when there is a control group.

 Deviation. Compares the mean of each level (except a reference category) to the

mean of all of the levels (grand mean). The levels of the factor can be in any order.

 Difference. Compares the mean of each level (except the first) to the mean of

previous levels. They are sometimes called reverse Helmert contrasts.

 Helmert. Compares the mean of each level of the factor (except the last) to the

mean of subsequent levels.

Page 85



Repeated. Compares the mean of each level (except the last) to the mean of

the subsequ

 Polynomial

The first degree of freedom contains the linear effect across all categories; the second degree of freedom, the quadratic effect; and so on. These contrasts are oftenusedt

Reference C

or factor level against which the others are compared.

Complex Sa

Figure 9-6

General Linear Model Save dialog box

Complex Sample

ent level.

Compares the linear effect, quadratic effect, cubic effect, and so on.

o estimate polynomial trends.

ategory.

The simple and deviation contrasts require a reference category,

mples General Linear Model Save

s General Linear Model

Save Variables. This group allows you to save the model predicted values and

residuals as new variables in the working file.

Page 86

Chapter 9

Export Model as SPSS data. Writes an SPSS data file containing a covariance (or

correlation, if selected) matrix of the parameter estimates in the model. Also, for each dependent v

ariable, there will be a row of p arameter estimates, a row of standard errors, a row of significance values for the t statistics corresponding to the parameter estimates, and a row of sampling design degrees of freedom. You can use this matrix file in oth

er procedures that read an SPSS matrix file.

Export Mod

el as XML.

Saves the parameter estimates and the parameter covariance matrix, if selected, in XML (PMML) format. SmartScore and the server version of SPSS (a separate product) can use this model file to apply the model information to other da

ta files for scoring purposes.

Complex Samples General Linear Model Options

Figure 9-7

General Linear Model Options dialog box

Missing Values.

User-

any covariates, must have valid data. Cases with invalid data for any of these variables are deleted from the analysis. These controls allow you to decide whether

missing values are treated as valid among the strata, cluster, subpopulation, and

userfactor variables.

All design variables, as well as the dependent variable and

Confidence Interval. This is the confidence interval level for coefficient estimates

and estimated marginal means. Specify a value greater than or equal to 50 and less

100.

than

Page 87

CSGLM Comman

The SPSS com

 Specify cus

(using the

 Fix covariates at values other than their means when computing estimated

marginal means (using the

 Specify a metric for polynomial contrasts (using the EMMEANS subcommand).  Specify a tolerance value for checking singularity (using the CRITERIA

subcommand

 Create use  Produce a g

See the SPSS Command Syntax Reference for complete syntax information.

Complex Sample

s General Linear Model

d Additional Features

mand language also allows you to:

tom tests of effects versus a linear combination of effects or a value

CUSTOM subcommand).

EMMEANS subcommand).

r-specified names for saved variables (using the

eneral estimable function table (using the

SAVE subcommand).

PRINT subcommand).

Page 88

Page 89

Chapter

Complex Samp

les

Logistic Regression

The Comple analysis on a binary or multinomial dependent variable for samples drawn by complex sampling methods. Optionally, you can request analyses for a subpopulation.

Example. A

different branches, according to a complex design. While incorporating the sample design, the officer wants to see if the probability with which a customer defaults is related t

Statistics. The procedure produces estimates, exponentiated estimates, standard

errors, confidence intervals, t tests, design effects, and square roots of design effects for model estimates. Pseudo R dependent and independent variables are also available.

Data. Th

are quantitative variables that are related to the dependent variable. Subpopulation variables can be string or numeric but should be categorical.

x Samples Logistic Regression procedure performs logistic regression

loan officer has collected past records of customers given loans at several

o age, employment history, and amount of credit debt.

parameters, as well as the correlations and covariances between parameter

statistics, classification tables, and descriptive statistics for the

e dependent variable is categorical. Factors are categorical. Covariates

ions.

Assumpt

that should be analyzed according to the specifications in the file selected in the Complex Samples Plan dialog box.

Obtaining Complex Samples Logistic Regression

From the menus choose:

Analyze

Complex Samples

Logistic Regression...

Thecasesinthedatafilerepresentasamplefromacomplexdesign

Page 90

Chapter 10

E Click Continue.

Select a plan file and, optionally, select a custom joint probabilities file.

Figure 10-1

Logistic Reg

ression dialog box

E Select a dependent variable.

Optionally, you can:

 Select variables for factors and covariates, as appropriate for your data.  Specify a variable to define a subpopulation. The analysis is performed only for

the selected category of the subpopulation variable.

Page 91

Complex Samp

Figure 10-2

Logistic Regression Reference Category dialog box

By default, the Complex Samples Logistic Regression procedure makes the highest-valued category the reference category. This dialog box allows you to specify the highest, lowest, or a custom category as the reference category.

Complex Sample

s Logistic Regression

les Logistic Regression Reference Category

Page 92

Chapter 10

Complex Samp

Figure 10-3

Logistic Regression Model dialog box

les Logistic Regression Model

Specify Model Effects. By default, the procedure builds a main-effects model using the

factors and covariates specified in the main dialog box. Alternatively, you can build a

om model that includes interaction effects and nested terms.

cust

Non-Nested Terms

For the selected factors and covariates:

raction.

Inte

Main effects. Creates a main-effects term for each variable selected.

Creates the highest-level interaction term for all selected variables.

Page 93

Complex Sample

All 2-way. Creates all possible two-way interactions of the selected variables. All 3-way. Creates all possible three-way interactions of the selected variables. All 4-way. Cr All 5-way. Creates all possible five-way interactions of the selected variables.

Nested Terms

You can build

eates all possible four-way interactions of the selected variables.

nested terms for your model in this procedure. Nested terms are

s Logistic Regression

useful for modeling the effect of a factor or covariate whose values do not interact with the levels of another factor. For example, a grocery store chain may follow the spending ha

bits of its customers at several store locations. Since each customer

frequents only one of these locations, the Customer effect can be said to be nested within the Store location effect.

Additiona

lly, you can include interaction effects or add multiple levels of nesting

to the nested term.

Limitations. Nested terms have the following restrictions:

 All factors within an interaction must be unique. Thus, if A is a factor, then

specifyin

 All facto

g A*A is invalid.

rs within a nested effect must be unique. Thus, if A is a factor, then

specifying A(A) is invalid.

 No effect can be nested within a covariate. Thus, if A is a factor and X is a

covariate, then specifying A(X) is invalid.

Intercept. The intercept is usually included in the model. If you can assume the

data pass through the origin, you can exclude the intercept. Even if you include the intercep

t in the model, you can choose to suppress statistics related to it.

Page 94

Chapter 10

Complex Samp

Figure 10-4

Logistic Regression Statistics dialog box

Model Fit. Controls the displays of statistics that measure the overall model

performance.

 Pseudo R-square. The R

counterpart among logistic regression models. There are, instead, multiple measures that attempt to mimic the properties of the R

 Classification table. Displays the tabulated cross-classifications of the observed

category by the model-predicted category on the dependent variable.

les Logistic Regression Statistics

statistic from linear regression does not have an exact

statistic.

Parameters. This group allows you to control the display of statistics related to the

model parameters.

 Estimate. Displays estimates of the coefficients.  Exponentiated estimate. Displays the base of the natural logarithm raised to the

power of the estimates of the coefficients. While the estimate has nice properties for statistical testing, the exponentiated estimate, or exp(B), is easier to interpret.

Page 95

Complex Sample



Standard error. Displays the standard error for each coefficient estimate.

 Confidence interval. Displays a confidence interval for each coefficient estimate.

The confiden

 t-test. Dis

ce level for the interval is set in the Options dialog box.

plays a t test of each coefficient estimate. The null hypothesis for each

s Logistic Regression

test is that the value of the coefficient is 0.

 Covariances of parameter estimates. Displays an estimate of the covariance matrix

for the model coefficients.

 Correlations of parameter estimates. Displays an estimate of the correlation matrix

for the mode

 Design eff

l coefficients.

ect.

The ratio of the variance of the estimate to the variance obtained by assuming that the sample is a simple random sample. This is a measure of the effect of specifying a complex design, where values further from 1 indicate greater ef

 Square ro

fects.

ot of design effect.

This is a measure of the effect of specifying a complex

design, where values further from 1 indicate greater effects.

Summary statistics for model variables. Displays summary information about the

dependent

variable, covariates, and factors.

Sample des

ign information.

Displays summary information about the sample,

including the unweighted count and the population size.

Page 96

Chapter 10

Complex Samp

Figure 10-5

Hypothesis Tests d ialog box

Test Statistic. This group allows you to select the type of statistic used for testing

hypotheses. You can choose between F, adjusted F, chi-square, and adjusted chi-square.

les Hypothesis Tests

Sampling Degrees of Freedom. This group gives you control over the sampling design

degrees of freedom used to compute p values for all test statistics. If based on the sampling design, the value is the difference between the number of primary sampling units and the number of strata in the first stage of sampling. Alternatively, you can set a custom degrees of freedom by specifying a positive integer.

Adjustment for Multiple Comparisons. When performing many hypothesis tests, you

run the risk of increased overall Type I error (the probability of incorrectly rejecting a null hypothesis). This group allows you to choose the method of adjusting the significance level.

Page 97

Complex Sample



Least significant difference. This method does not control the overall probability

of rejectin

g the hypotheses that some linear contrasts are different from the

null hypothesis values.

 Sequential Sidak. This is a sequentially step-down rejective Sidak procedure

that is much less conservative in terms of rejecting individual hypotheses but maintains t

 Sequentia

he same overall significance level.

l Bonferroni.

This is a sequentially step-down rejective Bonferroni procedure that is much less conservative in terms of rejecting individual hypotheses but maintaining the same overall significance level.

 Sidak. This method provides tighter bounds than the Bonferroni approach.  Bonferroni. This method adjusts the observed significance level for the fact that

multiple c

ontrasts are being tested.

Complex Samples Logistic Regression Odds Ratios

Figure 10-6

Logistic Regression Odds Ratios dialog box

s Logistic Regression

Page 98

Chapter 10

The Odds Ratios dialog box allows you to display the model-estimated odds ratios for specified factors and covariates. A separate set of odds ratios is computed for each category of

the dependent variable except the reference category.

Factors. Fo

r each selected factor, displays the ratio of the odds at each category of the

factor to the odds at the specified reference category.

Covariates. For each selected covariate, displays the ratio of the odds at the covariate’s

mean value plus the specified units of change to the odds at the mean.

When computing odds ratios for a factor or covariate, the procedure fixes all other factors at

their highest levels and all other covariates at their means. If a factor or

covariate interacts with other predictors in the model, then the odds ratios depend not only on the change in the specified variable but also on the values of the variables with whic

h it interacts. If a specified covariate interacts with itself in the model (for

example, age*age), then the odds ratios depend on both the change in the covariate and the value of the covariate.

Page 99

Complex Samp

Figure 10-7

Logistic Regression Save dialog box

Complex Sample

les Logistic Regression Save

s Logistic Regression

Save Variables. This group allows you to save the model-predicted category and

predicted probabilities as new variables in the working file.

Export Model as SPSS data. Writes an SPSS data file containing a covariance (or

correlation, if selected) matrix of the parameter estimates in the model. Also, for each dependent variable, there will be a row of parameter estimates, a row of standard errors, a row of significance values for the t statistics corresponding to the parameter estimates, and a row of sampling design degrees of freedom. You can use this matrix file in other procedures that read an SPSS matrix file.

Page 100

Chapter 10

Export Model as XML. Saves the parameter estimates and the parameter covariance

matrix, if selected, in XML (PMML) format. SmartScore and the server version of SPSS (a sepa to other data files for scoring purposes.

rate product) can use this model file to apply the model information

Complex Sam

Figure 10-8

Logistic Regression Options dialog box

ples Logistic Regression Options

Estimation. This group gives you control of various criteria used in the model

estimation.

 Maximum Iterations. The maximum number of iterations the algorithm will

execute. Specify a non-negative integer.

 Maximum Step-Halving. At each iteration, the step size is reduced by a factor

5 until the log-likelihood increases or maximum step-halving is reached.

of 0. Specify a positive integer.

Spss COMPLEX SAMPLES 13.0 User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Properties of Complex Samples

Usage of Complex Samples Procedures

Plan Files

Tree Controls for Navigating the Sampling Wizard

Sampling Wizard: Finish

CSPLAN and CSSELECT Commands Additional Features

Creating a New Analysis Plan

Tree Controls for Navigating the Analysis Wizard

Define Unequal Sizes

Analysis Preparation Wizard: Finish

Complex Samples Missing Values

Complex Samples Options

Complex Samples Options

Complex Samples Options

Complex Samples Hypothesis Tests

Complex Samples General Linear Model Options

Complex Samples Logistic Regression Odds Ratios