IBM Water System SPSS COMPLEX SAMPLES 19 Instrukcja Użytkownika Strona 1

iIBM SPSS Complex Samples 19

18 Complex Samples Ratios 171UsingComplexSamplesRatiostoAidPropertyValueAssessment... 171RunningtheAnalysis...

86Chapter 12Sequential Bonferroni. This is a sequentially step-down rejective Bonferroni procedure that ismuch less conservative in terms of rejectin

87Complex Samples Cox RegressionUpper bound of confidence interval for survival function. Saves the upper bound of the confidenceinterval for the surv

88Chapter 12Names o f Saved Variables. Automatic name generation ensures that you keep all your work.Custom names allow you to discard/replace results

89Complex Samples Cox Regressionvarname_. Takes values P1, P2, ..., corresponding to an ordered list of all model parameters,for row types COV or COR

90Chapter 12OptionsFigure 12-12Cox Regression dialog box, Options tabEstimation. These controls specify criteria for estimation of regression coefficie

91Complex Samples Cox Regressioniteration (the initial estimates), where n is the value of the increment. If the iteration history isrequested, then t

Part II:Examples

Chapter13Complex S amples Sampling WizardThe Sampling Wizard guides you through the steps for creating, modifying, or executing asampling plan file. Be

94Chapter 13Figure 13-1Sampling Wizard, Welcome stepE Select Design a sample,browsetowhereyouwanttosavethefile, and type property_assess.csplanas the n

95Complex Samples Sampling WizardFigure 13-2Sampling Wizard, Design Variables step (stage 1)E Select County as a stratification variable.E Select Towns

ParameterEstimates... 201Classification...

96Chapter 13Figure 13-3Sampling Wizard, Sample Size ste p (stage 1)E Select Counts from the U nits drop-down list.E Type 4 as the value for the number

97Complex Samples Sampling WizardFigure 13-4Sampling Wizard, Plan Summary step ( stage 1)E Select Yes,addstage2now.E Click Next.

98Chapter 13Figure 13-5Sampling Wizard, Design Variables step (stage 2)E Select Neighborhood as a stratification variable.E Click Next, and then click

99Complex Samples Sampling WizardFigure 13-6Sampling Wizard, Sample Size ste p (stage 2)E Select Proportions from the Units drop-down list.E Type 0.2

100Chapter 13Figure 13-7Sampling Wizard, Plan Summary step ( stage 2)E Look over the sampling design, and then click Next.

101Complex Samples Sampling WizardFigure 13-8Sampling W izard, Draw Sample, Selection Options stepE Select Custom value for the type of random seed to

102Chapter 13Figure 13-9Sampling Wizard, Finish stepE Click Finish.These selections produce the sampling plan file property_assess.csplan and draw a sa

103Complex Samples Sampling WizardPlan SummaryFigure 13-10Plan summaryThe summary table reviews your sampling plan and is useful for making sure that

104Chapter 13Figure 13-12Stage summaryThis summary table (the top part of which is shown here) reviews the second stage of sampling.It is also useful

105Complex Samples Sampling WizardThe agency will now use its resources to collect current valuations for t he properties selected inthe sample. Once

AppendicesA Sample Files 258B Notices 267Bibliography 269Index 271xii

106Chapter 13Figure 13-14Sampling Wizard, Welcome stepE Select Designasample, browse to where you want to save the file, and type demo.csplan asthename

107Complex Samples Sampling WizardFigure 13-15Sampling Wizard, Design Variables step (stage 1)E Select Region as a stratification variable.E Select Pro

108Chapter 13Figure 13-16Sampling Wizard, Sample Size ste p (stage 1)E Select Counts from the U nits drop-down list.E Type 3 as the value for the numb

109Complex Samples Sampling WizardFigure 13-17Sampling Wizard, Plan Summary step ( stage 1)E Select Yes,addstage2now.E Click Next.

110Chapter 13Figure 13-18Sampling Wizard, Design Variables step (stage 2)E Select District as a stratification variable.E Select City as a cluster vari

111Complex Samples Sampling WizardFigure 13-19Sampling Wizard, Sample Size ste p (stage 2)E Select Proportions from the Units drop-down list.E Type 0.

112Chapter 13Figure 13-20Sampling Wizard, Plan Summary step ( stage 2)E Select Yes,addstage3now.E Click Next.

11 3Complex Samples Sampling WizardFigure 13-21Sampling Wizard, Design Variables step (stage 3)E Select Subdivision as a stratification variable.E Clic

114Chapter 13Figure 13-22Sampling Wizard, Sample Size ste p (stage 3)E Select Proportions from the Units drop-down list.E Type 0.2 as the value for th

11 5Complex Samples Sampling WizardFigure 13-23Sampling Wizard, Plan Summary step ( stage 3)E Look over the sampling design, and then click Next.

Part I:User’s Guide

116Chapter 13Figure 13-24Sampling Wizard, Draw Sample Selection Options stepE Select 1, 2 as the stages to sample now.E Select Custom value for the ty

11 7Complex Samples Sampling WizardFigure 13-25Sampling Wizard, Finish stepE Click Finish.These selections produce the sampling plan file demo.csplan a

118Chapter 13Sample ResultsFigure 13-26Dat a E dito r with sample resultsYou can seethe sampling results in the Data Editor. Five new variables were s

119Complex Samples Sampling WizardFigure 13-27Sampling Wizard, Welcome stepE Select Draw a sample, browse to where you saved the plan file, and select

120Chapter 13Figure 13-28Sampling Wizard, Plan Summary step ( stage 3)E Select 1, 2 as stages already sampled.E Click Next.

121Complex Samples Sampling WizardFigure 13-29Sampling Wizard, Draw Sample Selection Options stepE SelectCustom value for the type of random seed to u

122Chapter 13Figure 13-30Sampling Wizard, Finish stepE Select Paste the syntax generated by the Wizard into a syntax window.E Click Finish.The followi

123Complex Samples Sampling WizardSample ResultsFigure 13-31Dat a E dito r with sample resultsYou can see the sampling results in the Data Editor. Thr

124Chapter 13Figure 13-32Sampling Wizard, Welcome stepE Select Designasample,browsetowhereyouwanttosavethefile, and type poll.csplan as thename of the

125Complex Samples Sampling WizardFigure 13-33Sampling Wizard, Design Variables step (stage 1)E Select County as a stratification variable.E Select Tow

126Chapter 13Figure 13-34SamplingWizard,SamplingMethodstep(stage1)E Select PPS as the sampling method.E Select Count data records as the measure of si

127Complex Samples Sampling WizardFigure 13-35Sampling Wizard, Sample Size ste p (stage 1)E Select Proportions from the Units drop-down list.E Type 0.

128Chapter 13Figure 13-36Sampling Wizard, Plan Summary step ( stage 1)E Select Yes,addstage2now.E Click Next.

129Complex Samples Sampling WizardFigure 13-37Sampling Wizard, Design Variables step (stage 2)E Select Neighborhood as a stratification variable.E Clic

130Chapter 13Figure 13-38Sampling Wizard, Sample Size ste p (stage 2)E Select Proportions from the Units drop-down list.E Type 0.2 as the value of the

131Complex Samples Sampling WizardFigure 13-39Sampling Wizard, Plan Summary step ( stage 2)E Look over the sampling design, and then click Next.

132Chapter 13Figure 13-40Sampling Wizard, Draw Sample Selection Options stepE Select Custom value for the type of random seed to use, and type 592004

133Complex Samples Sampling WizardFigure 13-41Sampling Wizard, Draw Sample Selection Options stepE Choose to save the sample to a new dataset, and typ

134Chapter 13Figure 13-42Sampling Wizard, Finish stepE Click Finish.These selections produce the sampling plan file poll.csplan anddrawasampleaccording

135Complex Samples Sampling WizardPlan SummaryFigure 13-43Plan summaryThe summary table reviews your sampling plan and is useful for making sure that

Chapter1Introduction to Complex SamplesProceduresAn inherent assumption of analytical procedures in traditional software packages is that theobservati

136Chapter 13Figure 13-45Stage summaryThis summary table (the top part of which is shown here) reviews the second stage of sampling.It is also useful

137Complex Samples Sampling WizardSample ResultsFigure 13-46Dat a E dito r with sample resultsYou can see the sampling results in the newly created da

138Chapter 13Figure 13-47Dat a E dito r with sample resultsUnlike voters in the second stage, the first-stage sampling weights are not identical for to

139Complex Samples Sampling Wizardinclusion probability matrices are 4×4 for these strata, and the Joint_Prob_5_ column is left emptyfor these rows. S

Chapter14Complex Samples AnalysisPreparation WizardThe Analysis Preparation Wizard guides you through the steps for creating or modifying ananalysis p

141Complex Samples Analysis Preparation WizardFigure 14-1Analysis Preparation Wizard, Welcome stepE Browse to where you want to save the plan file and

142Chapter 14Figure 14-2Analysis Preparation W izard, Design Variables step (stage 1)The data are obtained using a complex multistage sample. However,

143Complex Samples Analysis Preparation WizardSummaryFigure 14-3SummaryThe summary table reviews your analysis plan. The plan consists of one stage wi

144Chapter 14Figure 14-4Compute Variable dialog boxFifteenout of one hundred bank branches were selected without replacement in the first stage;thus, t

145Complex Samples Analysis Preparation WizardFigure 14-5Compute Variable dialog boxOne hundred customers were selected from each branch in the second

2Chapter 1Unequal selection probabilities. When sampling clusters that contain unequal numbers of units,you can use probability-proportional-to-size (

146Chapter 14Figure 14-6Compute Variable dialog boxNow that you have the inclusion probabilities for each stage, it’s easy to compute the finalsampling

147Complex Samples Analysis Preparation WizardFigure 14-7Analysis Preparation Wizard, Welcome stepE Browse to where you want to save the plan file and

148Chapter 14Figure 14-8Analysis Preparation W izard, Design Variables step (stage 1)E Select Branch as a cluster variable.E Select finalweight as the

149Complex Samples Analysis Preparation WizardFigure 14-9Analysis Preparation Wizard, Estimation Method step (stage 1)E Select Equal WOR as the first-s

150Chapter 14Figure 14-10Analysis Preparation Wizard, Size step (stage 1)E Select Read values from variable and select inclprob_s1 as the variable con

151Complex Samples Analysis Preparation WizardFigure 14-11Analysis Preparation Wizard, Plan Summary step (stage 1)E Select Yes,addstage2now.E Click Ne

152Chapter 14Figure 14-12Analysis Preparation Wizard, Estimation Method step (stage 2)E Select Equal WOR as the second-stage estimation method.E Click

153Complex Samples Analysis Preparation WizardFigure 14-13Analysis Preparation Wizard, Size step (stage 2)E Select Read values from variable and selec

154Chapter 14SummaryFigure 14-14Summary tableThe summary table reviews your analysis plan. The plan consists of t wo stages with a designof one cluste

Chapter15Complex Samples FrequenciesThe Complex Samples Frequencies procedure produces frequency tables for selected variablesand displays univariate

3Introduction to Complex Samples ProceduresAn analyst who doesn’t have access to the sampling plan file can specify an analysis plan andrefer to that

156Chapter 15Figure 15-1Complex Samples Plan dialog boxE Browse to and select nhis2000_subset.csaplan. For more information, see the topic Sample File

157Complex Samples FrequenciesFigure 15-2Freq uencies dialog boxE Select Vitamin/mineral supplmnts-past 12 m as a frequency variable.E Select Age cate

158Chapter 15Frequency TableFigure 15-4Frequency table for variable/situationEach selected statistic is computed for each selected cell measure. The fi

159Complex Samples FrequenciesSummaryUsing t he Complex Samples Frequencies procedure, you have obtained statistics for the use ofnutritional suppleme

Chapter16Complex Samples DescriptivesThe Complex Samples Descriptives procedure d isplays univariate summary statistics for severalvariables. Optional

161Complex Samples DescriptivesFigure 16-1Complex Samples Plan dialog boxE Browse to and select nhis2000_subset.csaplan. For more information, see the

162Chapter 16Figure 16-2Descriptives dialog boxE Select Freq vigorous activity (times per wk) through Freq strength activity (times per wk) asmeasure

163Complex Samples DescriptivesEClick Continue.E Click OK in the Complex Samples Descriptives dialog box.Univariate StatisticsFigure 16-4Univariate st

164Chapter 16Each selected statistic is computed for each measure variable by values o f Age category.Thefirstcolumn contains estimates of the average

Chapter17Complex Samples CrosstabsThe Complex Samples Crosstabs procedure produces crosstabulation tables for pairs of selectedvariables and displays

Chapter2Sampling from a Complex DesignFigure 2-1Sampling Wizard, Welcome stepThe Sampling Wizard guides you through the steps for creating, modifying,

166Chapter 17Figure 17-1Complex Samples Plan dialog boxE Browse to and select demo.csplan. For more information, see the topic Sample Files inAppendix

167Complex Samples CrosstabsFigure 17-2Crosstabs d ialog boxE Select Newspaper subscription as a row variable.E Select Response as a column variable.E

168Chapter 17Figure 17-3Crosstabs Statistics dialog boxE Deselect Population size and select Row percent in the Cells group.E Select Odds ratio and Re

169Complex Samples CrosstabsRisk EstimateFigure 17-5Risk estimate for newspaper subscription by responseThe relative risk is a ratio o f event probabi

170Chapter 17Risk Estimate by SubpopulationFigure 17-6Risk estimate for newspaper subscription by response, controlling for income categoryRelative ri

Chapter18Complex Samples RatiosThe Complex Samples Ratios procedure displays univariate summary statistics for ratios ofvariables. Optionally, you can

172Chapter 18Figure 18-1Complex Samples Plan dialog boxE Browse to and select property_assess.csplan. For more information, see the topic Sample Files

173Complex Samples RatiosFigure 18-2Ratios dialog boxE Select Current value as a numerator variable.E Select Value at last appraisal as the denominato

174Chapter 18RatiosFigure 18-4Ratios tableThe default display of the table is very wide, so you will need to pivot it for a better view.Pivoting the R

175Complex Samples RatiosSome of the confidence intervals do not overlap; thus, you can conclude that the ratios for theWestern county are higher than

5Sampling from a Complex DesignEClick Next to continue through the Wizard.E Optionally, in the Design Variables step, you can define strata, clusters,

Chapter19Complex Samples General LinearModelThe Complex Samples General Linear Model (CSGLM) procedure performs linear regressionanalysis, as well as

177Complex S amples General Linear ModelFigure 19-1Complex Samples Plan dialog boxE Browse to and select grocery.csplan. For more information, see the

178Chapter 19Figure 19-2General Linear Model dialog boxE Select Amount spent as the dependent variable.E Select Who shopping for and Use coupons as fa

179Complex S amples General Linear ModelFigure 19-3Model dialog boxE Choose to build a Custom model.E Select Main effects as the type of term to build

180Chapter 19Figure 19-4General Linear Model Statistics dialog boxE Select Estimate, Standa rd error, Confidence interval,andDesign effect in the Mode

181Complex S amples General Linear ModelESelect a Simple contrast and 3 Self and family as the reference category for shopfor. Note that, onceselected

182Chapter 19Parameter EstimatesFigure 19-8Parameter estimatesThe parameter estimates show the effect of each predictor on Amount spent. The value of

183Complex S amples General Linear ModelThe parameter estimates are useful for quantifying the effect of each model term, but the estimatedmarginal me

184Chapter 19Figure 19-11Overall test resul ts for estimated marginal means of genderThe overall test table reports the results of a test of all of th

185Complex S amples General Linear ModelFigure 19-15Estimated marginal means by l evels of gender by s hopping styleThis table displays the model-esti

Note: Before using this information and the product it supports, read the general informationunder Notices on p. 267.This document contains proprietar

6Chapter 2Sampling Wizard: Design VariablesFigure 2-2Sampling Wizard, Design Variables stepThis step allows you to select stratification and clustering

Chapter20Complex Samples Logistic RegressionThe Complex Samples Logistic Regression procedure performs logistic regression analysis ona binary or mult

187Complex Samples Logistic RegressionFigure 20-1Complex Samples Plan dialog boxE Browse to and select bankloan.csaplan. For more information, see the

188Chapter 20Figure 20-2Logistic Regression dialog boxE Select Previously defaulted as the dependent variable.E Select Level of education as a factor.

189Complex Samples Logistic RegressionFigure 20-3Logistic Regression Reference Category dialog boxE Select Lowest value as the reference category.This

190Chapter 20Figure 20-5Logistic Regression Odds Ratios dialog boxE Choose to create odds ratios for the factor ed and the covariates employ and debti

191Complex Samples Logistic RegressionWhat constitutes a “good” R2value varies between different areas of application. While thesestatistics can be su

192Chapter 20Each term in the model, plus the model as a whole, is tested for whether its effect equals 0. Termswith significance values less than 0.05

193Complex Samples Logistic RegressionOdds RatiosFigure 20-10Odds ratios for level of educationThis table displays the odds ratios of Previously defau

194Chapter 20This table displays the odds ratio of Previously defaulted for a unit change in the covariate Debt toincome ratio. The reported value is

Chapter21Complex Samples Ordinal RegressionThe Complex Samples Ordinal Regression procedure creates a predictive model for an ordinaldependent variabl

7Sampling from a Complex DesignInput Sample Weight. If the current sample design is part of a larger sample design, you may havesample weights from a

196Chapter 21Figure 21-1Complex Samples Plan dialog boxE Browse to and select poll.csplan as the plan file. For more information, see the topic Sample

197Complex Samples Ordinal RegressionFigure 21-2Ordinal Regression dialog boxE Select The legislature should enact a gas tax as the dependent variable

198Chapter 21Figure 21-3Ordinal Regression Statistics dialog boxE Select Classification table in the M odel Fit group.E Select Estimate, Exponentiated

199Complex Samples Ordinal RegressionFigure 21-4Hypothesis Tests dialo g b oxEven for a moderate number of predictors and response categories, the Wal

20 0Chapter 21EClick OK in the Complex Samples Ordinal Regression dialog box.Pseudo R-SquaresFigure 21-6Pseudo R-Sq uaresIn the linear regression mode

201Complex Samples Ordinal RegressionEach term in the model is tested for whether its effect equals 0. Terms with significance valuesless than 0.05 hav

202Chapter 21Those who drive less frequently show greater support for the bill than those who drive morefrequently. The coefficients for the variable

203Complex Samples Ordinal RegressionFigure 21-10Classification tableThe classification table shows the practical results of using the model. For each

204Chapter 21merely the ratios of the exponentiated parameter estimates. For example, the cumulative oddsratio for 18–30 vs. >60 is 1.00/0.723 = 1.

205Complex Samples Ordinal RegressionFigure 21-14Parameter estimates for generalized cumulative model (shown in part)Moreover, the estimated values of

8Chapter 2Sampling Wizard: Sampling MethodFigure 2-3Sampling Wizard, Sampling Method stepThis step allows you to specify how to select cases from the

206Chapter 21EClick Continue in the Plan dialog box.Figure 21-15Ordinal Regression dialog boxE Deselect Gender and Voted in last election as factors.E

207Complex Samples Ordinal RegressionFigure 21-16Ordinal Regression Options dialog boxE Select Display iteration history.The iteration history is usef

208Chapter 21Figure 21-18Warnings for reduced modelLooking at the iteration history, the changes in the parameter estimates over the last few iteratio

209Complex Samples Ordinal Regressionfrom Disagree to Agree, more than half of whom were observed to respond Disagree or Stronglydisagree. This is a v

Chapter22Complex Samples Cox RegressionThe Complex Samples Cox Regression procedure performs survival analysis for samples drawnby complex sampling me

211Complex Samples Cox RegressionFigure 22-1Compute Variable dialog boxE Type date2as the target variable.E Type DATE.DMY(30,6,2006)as the expression.

212Chapter 22Figure 22-2ComputeVariableIfCasesdialogboxE Select Include if case satisfies condition.E Type MISSING(date2) as the expression.E Click Co

213Complex Samples Cox RegressionFigure 22-3DateandTimeWizard,WelcomestepE Select Calculate with dates and time s.E Click Next.

214Chapter 22Figure 22-4Date and T ime Wizard, Do Calculations on Dates stepE Select Calculate the number of time units between two dates.E Click Next

215Complex Samples Cox RegressionFigure 22-5Date and Time W izard, Calculate the nu mber of time units between two dates stepE Select Date of second a

9Sampling from a Complex DesignPPS Systematic. This is a first-stage method that systematically selects units with probabilityproportional to size. Th

216Chapter 22Figure 22-6Date and Time W izard, Calculation stepE Type time_to_event as the name o f the variable representing the time between the two

217Complex Samples Cox RegressionFigure 22-7Complex Samples Plan for Cox Regression dialog boxE Browse to the sample files directory and select recidiv

218Chapter 22Figure 22-8Cox Regression dialog box, Ti me and Event tabE Select Time to second arrest [time_to_event] asthevariabledefining the end of t

219Complex Samples Cox RegressionFigure 22-9Define Event dialog boxE Select 1Yesas the value indicating the event of interest (rearrest) has occurred.

220Chapter 22Figure 22-10Cox Regression dialog box, Predictors tabE Select Age in years [age] as a covariate.E Click the Statistics tab.

221Complex Samples Cox RegressionFigure 22-11Cox Regression dialog box, Statistics tabE Select Test of proportional hazards and then select Log as the

222Chapter 22This table contains information on the sample design pertinent to the estimation of the model. There is one case per subject, and all 5,

223Complex Samples Cox RegressionEClick New.Figure 22-16Cox Regression Define Time-Dependent Predictor dialog boxE Type t_age as the name of the time-

224Chapter 22Figure 22-17Cox Regression dialog box, Predictors tabE Select t_age as a covariate.E Click the Statistics tab.

225Complex Samples Cox RegressionFigure 22-18Cox Regression dialog box, Predictors tabE Select Estimate, Standard error, Confidence interval,andDesign

10Chapter 2Sampling Wizard: Sample SizeFigure 2-4Sampling Wizard, S a mple Size stepThis step allows you to specify the number or proportion of units

226Chapter 22Parameter EstimatesFigure 22-20Parameter estimatesLooking at the parameter estimates and standard errors, you can see that you have repli

227Complex Samples Cox RegressionPreparing the Data for AnalysisBefore restructuring the data, you will need to create two ancillary variables to help

228Chapter 22ERecall the Compute Variable dialog box.Figure 22-22Compute Variable dialog boxE Type start_time3 as the target variable.E Type time2 as

229Complex Samples Cox RegressionFigure 22-23Restru cture D ata W izard, Welcome stepE Make sure R estructure selected variables into cases is selecte

230Chapter 22Figure 22-24Restructure Data Wizard, Variables to Cases Num ber of Variable Groups stepE Select More than one variable group to restructu

231Complex Samples Cox RegressionFigure 22-25Restructure Dat a Wizard, Variables to Cases Select Variables stepE In the Case Group Identification group

232Chapter 22Figure 22-26Restructure Dat a Wizard, Variables to Cases Select Variables stepE Type start_time as the target variable.E Select Length of

233Complex Samples Cox RegressionFigure 22-27Restructure Dat a Wizard, Variables to Cases Select Variables stepE Type time_to_event as the target vari

234Chapter 22Figure 22-28Restructure Dat a Wizard, Variables to Cases Select Variables stepE Type mi as the target variable.E Select History of myocar

235Complex Samples Cox RegressionFigure 22-29Restructure Dat a Wizard, Variables to Cases Select Variables stepE Type is as the target variable.E Sele

11Sampling from a Complex DesignDefine Unequal SizesFigure 2-5Define Unequal Sizes dialog boxThe Define Unequal Sizes dialog box allows you to enter si

236Chapter 22Figure 22-30Restructure Dat a Wizard, Variables to Cases Select Variables stepE Type hs as the target variable.E Select History o f hemor

237Complex Samples Cox RegressionFigure 22-31Restru cture Data Wizard, Variables to Cases Create One Index Variable stepE Type event_index as the name

238Chapter 22Figure 22-32Restru cture Data Wizard, Variables to Cases Create One Index Variable stepE Make sure Keep and treat as fixed variable(s) is

239Complex Samples Cox RegressionFigure 22-33Restructured dataThe restructured data contains three cases for every patient; however, many p atients ex

240Chapter 22Figure 22-34Select Cases d ialog boxE SelectIf condition is satisfied.E Click If.

241Complex Samples Cox RegressionFigure 22-35Select Cases If d ialog boxE Type event >= 0 as the conditional expression.E Click Continue.

242Chapter 22Figure 22-36Select Cases d ialog boxE SelectDelete unselected cases.E Click OK.Creating a Simple Random Sampling Analys is PlanNow you ar

243Complex Samples Cox RegressionFigure 22-37Cox Regression main dialog boxE Type sampleweight as the target variable.E Type 1 as the numeric expressi

244Chapter 22Figure 22-38Analysis Preparation Wizard, Welcome stepE Select Createaplanfileand type srs.csaplan as the name of the file. Alternatively,

245Complex Samples Cox RegressionFigure 22-39Analysis Preparation Wizard, Design VariablesE Select sampleweight as the sample weight variable.E Click

12Chapter 2Sampling Wizard: Output VariablesFigure 2-6Sampling Wizard, O utput Variables stepThis step allows you to choose variables to save when the

246Chapter 22Figure 22-40Analysis Preparation Wizard, Estimation MethodE Deselect Use finite population corre ction.E Click Finish.You are now r eady

247Complex Samples Cox RegressionFigure 22-41Plan for Cox Regression dialog boxE Browse to where you saved the simple random sampling analysis plan, o

248Chapter 22Figure 22-42Cox Regression dialog box, Ti me and Event tabE Select Varies by subject and select Length of stay for rehabilitation [los_re

249Complex Samples Cox RegressionFigure 22-43Define Event dialog boxE Select 4Deathas the value indicating the terminal event has occurred.E Click Con

250Chapter 22Figure 22-44Cox Regression dialog box, Ti me and Event tabE Select Patient ID [patid] as the subject identifier.E Click the Predictors tab

251Complex Samples Cox RegressionFigure 22-45Cox Regression dialog box, Predictors tabE Select History of myocardial infarction [mi] through History o

252Chapter 22Figure 22-46Cox Regression dialog box, Statistics tabE Select Estimate, Exponentiated estimate, Standard error,andC onfidence interval in

253Complex Samples Cox RegressionFigure 22-47Cox Regression dialog box, Statistics tabE Select Log-minus-log of survival function.E Check Separate Lin

254Chapter 22Figure 22-48Cox Regression dialog box, Options tabE Select Breslow as the tie-breaking method in the Estimation group.E Click OK.Sample D

255Complex Samples Cox RegressionThere are multiple cases for some subjects, and all 3,310 cases are used in the analysis. The design has a single s

13Sampling from a Complex DesignIndex. Identifies units selected multiple times within a given stage. The rootname for the savedvariable is Index_.Note

256Chapter 22The confidence intervals for [mi=0] and [mi=1] do not overlap with the interval for [mi=2],and none of them include 0. Therefore, it appe

257Complex Samples Cox RegressionLog-Minus-Log PlotFigure 22-53Log-minus-log plotThis plot displays the log-minus-log of the survival function, ln(−ln

AppendixASample FilesThe sample files installed with the product can be found in the Samples subdirectory of theinstallation directory. There is a sepa

259Sample Filesloans. The last 150 cases are prospective customers that the bank needs t o classify as goodor bad credit risks. bankloan_binning.sav.

260Appendix Acarpet_prefs.sav. This data file is based on the same example as described for carpet.sav,butitcontains the actual rankings collected fro

261Sample Filesdebate.sav. This is a hypothetical data file that concerns paired responses to a survey fromattendees of a political debate before and

262Appendix Aguttman.sav. Bell (Bell, 1961) presented a table to illustrate possible social groups. Guttman(Guttman, 1968) used a portion of this tab

263Sample Filesnhis2000_subset.sav. The National Health Interview Survey (NHIS) is a large, population-basedsurvey of the U.S. civilian population. I

264Appendix Aand sample weights. The additional variable Current value was collected and added to thedata file after the sample was t aken. recidivism

265Sample Filesstroke_clean.sav. This hypothetical data file contains the state of a medical database after ithas been cleaned using procedures in the

14Chapter 2Sampling Wizard: Draw Sample Selection OptionsFigure 2-8Sampling Wizard, Draw Sample Selection Options stepThis step allows you to choose w

266Appendix Atv-survey.sav. This is a hypothetical data file that concerns a survey conducted by a TV studiothat is considering whether to extend the

AppendixBNoticesLicensed Materials – Property of SPSS Inc., an IBM Company. © Copyright SPSS Inc. 1989,2010.Patent No. 7,023,453The following paragrap

268Appendix Busing, marketing or distributing application programs conforming to the application programminginterface for the operating platform for w

BibliographyBell,E.H.1961. Social foundations of human behavior: Introduction to the study of sociology.New York: Harper & Row.Blake, C. L., and C

270BibliographyRosenberg, S., and M. P. Kim. 1975. The method of sorting as a data-gathering procedure inmultivariate research. Multivariate Behaviora

Indexadjusted chi-squarein Complex Samples, 49, 59, 69in Complex Samples Cox Regression, 85adjusted F statisticin Complex Samples, 49, 59 , 69in Compl

272Indexpseudo R2statistics, 190reference category, 55related procedures, 194save variables, 61statistics, 57tests of m odel effects, 1 91Complex Samp

273Indexiterationsin Comp lex Samples L ogistic Regression, 62in Complex Sample s Ordinal Regression, 72least signific ant differencein Complex Samples

274Indexsample weightsin Analysis Preparation Wizard, 20in Sampling Wizard, 12samplingcomplex design, 4sampling estimationin Analysis Preparation Wiza

15Sampling from a Complex DesignSampling Wizard: Draw Sam ple Output FilesFigure 2-9Sampling Wizard, Draw Sample Output Files stepThis step allows you

PrefaceIBM® SPSS® S tatistics is a comprehensive syste m for analyzing data. The Complex Samplesoptional add-on module provides the additional analyti

16Chapter 2Sampling Wizard: FinishFigure 2-10Sampling Wizard, Finish stepThis is the final step. You can save the plan file and draw the sample now or p

17Sampling from a Complex DesignEReviewthesamplingplaninthePlanSummarystep,andthenclickNext.Subsequent steps are largely the same as for a new design.

18Chapter 2Remove stages. You can remove stages 2 and 3 from a multistage design.Running an Existing Sample PlanE From the menus choose:Analyze > C

Chapter3Preparing a Complex Sample forAnalysisFigure 3-1Analysis Preparation Wizard, Welcome stepThe Analysis Preparation Wizard guides you through th

20Chapter 3Creating a New Analysis PlanE From the menus choose:Analyze > ComplexSamples > Prepare for Analysis...ESelect Createaplanfile,andchoo

21Preparing a Complex Sampl e for AnalysisThis step allows you to identify the stratification and clustering variables and define sampleweights. You can

22Chapter 3Analysis Preparation Wizard: Estimation MethodFigure 3-3Analysis Preparation Wizard, Estimation Method stepThis step allows you to specify

23Preparing a Complex Sampl e for AnalysisAnalysis Preparation W izard: SizeFigure 3-4Analysis Preparation Wizard, Size stepThis step is used to speci

24Chapter 3Define Unequal SizesFigure 3-5Define Unequal Sizes dialog boxThe Define Unequal Sizes dialog box allows you to enter sizes on a per-stratum

25Preparing a Complex Sampl e for AnalysisAnalysis Preparation Wizard: Plan SummaryFigure 3-6Analysis Preparation Wizard, Plan Summary stepThis is the

Additional PublicationsThe SPSS Statistics: Guide to Data Analysis, SPSS Statistics: Statistical Procedures Companion,and SPSS Statistics: Advanced St

26Chapter 3Analysis Preparation Wizard: FinishFigure 3-7Analysis Preparation Wizard, Finish stepThis is the final step. You can save the plan file now o

27Preparing a Complex Sampl e for AnalysisEReview the analysis plan in the Plan Summary step, and then click Next.Subsequent steps are largely the sam

Chapter4Complex Samples PlanComplex Samples analysis procedures require analysis specifications from an analysis or sampleplan file in order to provide

Chapter5Complex Samples FrequenciesThe Complex Samples Frequencies procedure produces frequency tables for selected variablesand displays univariate s

30Chapter 5Figure 5-1Freq uencies dialog boxE Select at least one frequency variable.Optionally, you can specify variables to define subpopulations. St

31Complex Samples FrequenciesConfidence interval. Aconfidence interval for the estimate, using the specified level. Coefficient of variation. The rati

32Chapter 5Complex Samples OptionsFigure 5-4Options dialog boxSubpopulation Display. You can choose to have subpopulations displayed in the same table

Chapter6Complex Samples DescriptivesThe Complex Samples Descriptives procedure d isplays univariate summary statistics for severalvariables. Optionall

34Chapter 6Figure 6-1Descriptives dialog boxE Select at least one measure variable.Optionally, you can specify variables to define subpopulations. Stat

35Complex Samples DescriptivesSummaries. This group allows you to request estimates of the means and sums of the measurevariables. Additionally, you c

ContentsPart I: User’s Guide1 Introduction to Complex Samples Procedures 1PropertiesofComplexSamples ...

36Chapter 6Complex Samples OptionsFigure 6-4Options dialog boxSubpopulation Display. You can choose to have subpopulations displayed in the same table

Chapter7Complex Samples CrosstabsThe Complex Samples Crosstabs procedure produces crosstabulation tables for pairs of selectedvariables and displays t

38Chapter 7Figure 7-1Crosstabs d ialog boxE Select at least one row variable and one column variable.Optionally, you can specify variables to define su

39Complex Samples CrosstabsComplex Samples Crosstabs StatisticsFigure 7-2Crosstabs Statistics dialog boxCells. This group allows you to request estima

40Chapter 7Summaries for 2-by-2 Tables. This group produces statistics for tables in which the row and c olumnvariable each have two categories. Each

41Complex Samples CrosstabsComplex Samples OptionsFigure 7-4Options dialog boxSubpopulation Display. You can choose to have subpopulations displayed i

Chapter8Complex Samples RatiosThe Complex Samples Ratios procedure displays univariate summary statistics for ratios ofvariables. Optionally, you can

43Complex Samples RatiosFigure 8-1Ratios dialog boxE Select at least one numerator variable and denominator variable.Optionally, you can specify varia

44Chapter 8Design effect. The ratio of the variance of the estimate to the variance obtained by assumingthat the sample is a simple random sample. Th

Chapter9Complex Samples General LinearModelThe Complex Samples General Linear Model (CSGLM) procedure performs linear regressionanalysis, as well as a

AnalysisPreparationWizard:EstimationMethod...22AnalysisPreparationWizard:Size ...

46Chapter 9Figure 9-1General Linear Model dialog boxE Select a dependent variable.Optionally, you can: Select variables for factors and covariates, a

47Complex S amples General Linear ModelFigure 9-2Model dialog boxSpecify Model Effects. By default, the procedure builds a main-effects model using th

48Chapter 9Nested TermsYou can build nested terms for your model in this procedure. Nested terms are useful for modelingthe effect of a factor or cova

49Complex S amples General Linear ModelCovariances of parameter estimates. Displays an estimate of the covariance matrix for themodel coefficients. C

50Chapter 9first stage of sampling. Alternatively, you can set a custom degrees of freedom by specifying apositive integer.Adjustment for Multiple Comp

51Complex S amples General Linear ModelThe Estimated Means dialog box allows you to display the model-estimated marginal means forlevels of factors an

52Chapter 9Export model as SPSS Statistics data. Writes a dataset in IBM® SPSS® Statistics format containingthe parameter correlation or covariance ma

53Complex S amples General Linear ModelCSGLM Command Additional FeaturesThe command syntax language also allows you to: Specify custom tests of effec

Chapter10Complex Samples Logistic RegressionThe Complex Samples Logistic Regression procedure performs logistic regression analysis ona binary or mult

55Complex Samples Logistic RegressionFigure 10-1Logistic Regression dialog boxE Select a dependent variable.Optionally, you can: Select variables for

9 Complex Samples General Linear Model 45ComplexSamplesGeneralLinearModelStatistics...48ComplexSamplesHypothesisTests

56Chapter 10By default, the Complex Samples Logistic Regression procedure makes the highest-valuedcategory the reference category. This dialog box all

57Complex Samples Logistic RegressionAll 3-way. Creates all possible three-way interactions of the selected variables.All 4-way. Creates all possible

58Chapter 10Pseudo R-square. The R2statistic from linear regression does not have an exact counterpartamong logistic regression models. There are, in

59Complex Samples Logistic RegressionComplex Samples Hypothesis TestsFigure 10-5Hypothesis Tests dialo g b oxTe st Statistic. This group allows you to

60Chapter 10Complex Samples Logistic Regression Odds RatiosFigure 10-6Logistic Regression Odds Ratios dialog boxThe Odds Ratios dialog box allows you

61Complex Samples Logistic RegressionComplex Samples Logistic Regression SaveFigure 10-7Logistic Regression Save dialog boxSave Variables. This group

62Chapter 10Complex S amples Logistic Regression OptionsFigure 10-8Logistic Regression Options dialog boxEstimation. This group gives you control of v

63Complex Samples Logistic RegressionConfidence Interval. This is the confidence interval level for coefficient e stimates, exponentiatedcoefficient esti

Chapter11Complex Samples Ordinal RegressionThe Complex Samples Ordinal Regression procedure performs regression analysis on a binaryor ordinal depende

65Complex Samples Ordinal RegressionFigure 11-1Ordinal Regression dialog boxE Select a dependent variable.Optionally, you can: Select variables for f

Predictors ...78DefineTime-DependentPredictor...

66Chapter 11Complex Sam ples Ordinal Regression Response P robabilitiesFigure 11-2Ordinal Regression Response Probabilities dialog boxThe Response Pro

67Complex Samples Ordinal RegressionNon-Nested TermsFor the selected factors and covariates:Interaction. Creates the highest-level interaction term fo

68Chapter 11Complex S amples Ordinal Regression StatisticsFigure 11-4Ordinal Regression Statistics dialog boxModel Fit. Controls the display of statis

69Complex Samples Ordinal RegressionDesign effect. The ratio of the variance of the estimate to the variance obtained by assumingthat the sample is a

70Chapter 11Te st Statistic. This group allows you to select the t ype o f statistic used for testing hypotheses. Youcan choose between F,adjustedF, c

71Complex Samples Ordinal RegressionFactors. For each selected factor, displays the ratio of the cumulative odds at each category of thefactor to the

72Chapter 11rowtype_. Takes values (and value labels), COV (Covariances), CORR (Correlations), EST(Parameter estimates), SE (Standard errors), SIG (S

73Complex Samples Ordinal RegressionEstimation Method. You can select a parameter estimation method; choose b etweenNewton-Raphson, Fisher scoring, or

Chapter12Complex Samples Cox RegressionThe Complex Samples Cox Regression procedure performs survival analysis for samples drawnby complex sampling me

75Complex Samples Cox RegressionSubject Identifier. You can easily incorporate piecewise-constant, time-dependent predictors bysplitting the observati

PreparingforAnalysisWhenSamplingWeightsAreNotintheDataFile... 143Computing Inclusion Proba bilitie s and Sa mpling Weights . . . . . .

76Chapter 12Figure 12-1Cox Regression dialog box, Ti me and Event tabE Specify the survival time by selecting the entry and exit times from the study.

77Complex Samples Cox RegressionDefine EventFigure 12-2Define Event dialog boxSpecify the values that indicate a terminal event has occurred. Individ

78Chapter 12PredictorsFigure 12-3Cox Regression dialog box, Predictors tabThe Predictors tab allows you to specify the factors and covariates used to

79Complex Samples Cox RegressionDefine Time-Dependent PredictorFigure 12-4Cox Regression Define Time-Dependent Predictor dialog boxThe Define Time-Depe

80Chapter 12Note: If your segmented, time-dependent predictor is constant within segments, as in the bloodpressure example given above, it may be easi

81Complex Samples Cox RegressionModelFigure 12-6Cox Regression dialog box, Model tabSpecify Model Effects. By default, the procedure builds a main-eff

82Chapter 12Nested TermsYou can build nested terms for your model in this procedure. Nested terms are useful for modelingthe effect of a factor or cov

83Complex Samples Cox RegressionSample design information. Displays summary information about the sample, including theunweighted count and the popula

84Chapter 12PlotsFigure 12-8Cox Regression dialog box, Plots tabThe Plots tab allows you to request plots of the hazard function, survival function, l

85Complex Samples Cox RegressionHypothesis Te stsFigure 12-9Cox Regression dialog box, Hypothesis Tests tabTe st Statistic. This group allows you to s