Re: Quantitative methods - cumulative file for BSA 1983 - 2014

[John F Hall]

Sarah

 

Nice to find someone interested in my work. 

 

Unfortunately I’m not at liberty to distribute my cumulative file unless you have authorisation from UKDS.  Natcen need to approve the deposit of the cumulative file, but since I’ve used files downloaded from UKDS, I suspect it will be quicker to go via them or, since you are a registered user, ask them if I can send you the file direct via Dropbox.  I have already sent them the link to download the beta1 version.

 

I’m currently working on a beta2 version of the “mother” file using a table generated with syntax supplied by Jon Peck (ex-IBM-SPSS, now retired). 

 

dataset declare values.

oms select tables /if subtypes='Variable Values'

/destination outfile=values format=sav viewer=no.

display dict.

omsend.

 

dataset activate values.

sort cases by Label

 

Try it on your file(s) to see what comes up.

 

It generates a table with several thousand rows in which variable names appear as many times as they have different variable labels.

 

EnjoyCJB              2.0000   Agree

PubOwnSt           2.0000   Agree

SecSchl1              1.0000   Agree   strongly

SecSchl4              1.0000   Agree   strongly

EnjoyCJB              8.0000   Can't choose

PubOwnSt           1.0000   Agree   strongly

 

Basically I can use the table to identify variables with unique label properties (eg “Neither satisfied nor dissatisfied”) then copy the variable names into a separate (annotated) file, thence to a new SPSS syntax file.  This is routine and repetitive, but far quicker than attempting it variable by variable: Jon has saved me days if not weeks of work.  It’s not a job for novices or the faint-hearted: it needs to be done by someone dedicated, very experienced at working with SPSS files, who knows what he is doing and can instinctively spot problems and anomalies.

 

So far, I’ve only done half of the measurement levels: missing values will be next.  My syntax file looks like this, but will be much bigger.  I know I should do it in small doses, but . .

 

title 'Pulling out variables using value labels'.

 

/* Agree – Disagree (1-5) .

/*(3 = Neither)..variable  level.

VARIABLE LEVEL

SecSchl1 SecSchl4 SecSchl5 RelSup1 RelSup2

(ORDINAL).

 

/*(3 = Neither agree nor disagree.).

VARIABLE LEVEL

PropCRe Wk65I PENSEC govnosa2 govnocar polundst infpoli2 govright govself polpar1 polpar2 polpar3

bnlowp bnmove pencomp earlyret credplan borohard marvie16 marview6 uniaff unicar undwry gmlmarry

welfhelp morewelf unempjob sochelp dolefidl welffeet damlives proudwlf redistrb bigbusnn wealth richlaw

indust4 tradvals stifsent deathapp obey wronglaw censor SavFrRet InStuEc InStuCu InStuOc InStuCs

CycDang CCALowE CCACar CCAPLANE AffEngl AffLocl BuilAff BuilUnf SupDes IDMus IDEaster

IDBAsian VacOW DateOW RespOW PresOW NormOW VacPT AskPT LocalPT FamPT DatePT RespPT

PresPT NormPT VacFT AskFT LocalFT DateFT RespFT PresFT NormFT ASAStay Soeqsame

NatCitzn NatAshmd NatLike NatBest NatSupp NatSport NatPrdBr NatShrt

ForgRel1 ForgRel2 ForgRel4 ForgRel5 ForgRel6 MNNTDNMG FreeTrde GBFollow MNNTPowr

CitWorld SharTrad EthncAid Immigrt1 Immigrt2 Immigrt3 Immigrt4 Immigrt6 Citzshp3 Citzshp4 Citzshp5

EUFollow lowwage UniFibet UniDebts UniNWort carwalk2 carbus2 carbike2 plnallow plnterm plnenvt plnuppri

cartaxhi carallow carreduc carnod2 carenvdc ddnodrv ddnklmt specamsl specammo specamtm speedlim

speavesc mobdsafe mobddang mobdban mobdlaw WestLoth SamsxPr PerVLook LookAchi GirlWApp GovNoSay

VoteOnly GovComp LowWage2 WWRELCHD WWCHDSUF WWFAMSUF WANTHOME HWIFEFFL BOTHEARN

SEXROLE MARVIEW1 MARVIE11 MARVIE13 MARVIE10 lescpar2 gaycpar2 chdview2 chdview3 chdview9

chdvie10 chdvie11 chdvie12 FEMJOB hwcmty mwdiff machnowk fammencr chmsca chmscb chmscc chmscd

chmsce chmscf chmscg chmsch chmsci chmscj CliCar CliPlane CCASpe ASWork AfAccept AfSucces AfWrong

AfSupMem IqAccept IqSucc IqWrong IqSupMem ESCompND ESCompDi LivAp1 LivAp2 LivAp3 LivAp4 LivAp5

LivAp6 LivAp7 LoseTch VoteIntr PtyNMat2 QuitRule QuitBdJb MYRDONE2 MYRPOWR2 MYRSPK2 PCCRIME

PCPOLINF HSysImpr UseHCS LimHCS HCIneff AcNoCit AcDamHe DmHSuff EnvSuff GeneSuff PoorSuff

AltMSol AltMProm TrustDrs DrsDisc DrSkNGd DrCaEarn DrTelMis BnTarget BnComp BnSlow HlthGap DirectFl

RelTime RoadCap CARNODi2 DrugNoDr SpeCAvFS MigWorPS MigrStop MiHardWk CTAXREF2 InfChoic

RefForce HLExpert Wk65Mon OldWk65 OldWkFlx OldWkChg IncRedis Happ10yr Adve10yr Edu10yr Safe10yr

YPTime Noplygrd YPBehav2 GirlsBv2 YPRes2 LikeSR NeligSR GoStatSc GoStaQua ManSkill COwnPref

CManPref RetStan2 Wk65Flx PRIVENT INCDIFF SCIEBELF SCIEHARM SCIESOLV FUTENVIR HARMEVIR

HARMVIRW ENVIRECG GROWHARM POPGROW ENVIRDIF ENVIRRGT MORIMPEN NOPTENV EXAGENV

DKHrmEnv EnvEfLif INTAGENV POORLESS ECPSLOW SOCWCHLD ChldGrow Ownhlth1 Ownhlth2 Ownhlth3

Ownhlth4 RelyDoc BRPyMr BTPyMr BTNoAlt CmplChrg MOTORWAY BUILDTRA CARNODIF DD5YBAN SpeCmInc

HOMERISK buycheap movehome moneytie freedom finburdn leavefam homeresp riskjob WAITFAM ScChRt

BestSch clsSchCm clsSchSC SchSlfsh UniAnyAg DegGdJob UniAdvt PREGTN11 SafeJob ManRelKp ManRelUn

ManRelHo ConfSc1 ConfSc2 ConfSc3 ConfSc4 ConfSc5 ConfSc6 ConfSc7 ConfSc8 ConfSc9 PropRep

ETHTAKJ2 EETAKJ2 GPDay GPEve GPOnly WhRchPar IncNec IncInev IncUnf IncWrk IncWrng Incprob

Bonus1 Bonus2 Bonus3 ScotPayE CREDSPND INCOPP1 INCOPP2 INCOPP3 EQUALOPP EQUALOP2 OldJobOp

WellInf EnjoyPdW EnjoyCJB PatFeel1 PatFeel2 PatFeel3 PatFeel4 GETON3 GETON4 GETON5 GETON6

INEQRICH INCDIFFS UNEMPSOL LESSBENF DisNtEff DisNoAll OldAlw PubOwnSt

(ORDINAL).

 

/*Satisfied - Dissatified (1-7)  .

VARIABLE LEVEL

BrHSSat  DrTrSat  AltMeSat  HspTrSat  jobsat2

famsat2  BrHSSat  DrTrSat  AltMeSat  HspTrSat

(ORDINAL).

 

/*Satisfied - Dissatified (1-5)  .

VARIABLE LEVEL

NHSSat GPSat DentSat InpatSat OutpaSat AESat CareSat3 Appearsa CareSat2

jobsat2 famsat2 NDirSat Ambsat diogsat AreaSat JobSat3 PaySatis HrsSatis PrDenSat

(ORDINAL).

 

/*SAME - WORSE.

VARIABLE LEVEL

NHS5yrs  Say5Yrs  NHSNx5Yr  WtOp5Yrs  WtAp5Yrs  WtCo5Yrs  WtGP5Yrs  WtRf5Yrs  Des20yr 

PrejNow  PrejFut  UniJobs  AgeDNow  AgeDOldW  AgeDYngW   

(ORDINAL).

 

/*Approve – disapprove.

VARIABLE LEVEL

PHelpSc1 PHelpSc2 PHelpSc3 PHelpSc4 PHelpSc5 PHelpSc6

((ORDINAL).

 

/*Comfortable - Uncomfortable.

VARIABLE LEVEL

HIncDif4 FlexAsk TalkLGB SocLGB

DMPSens  DMPPhys  DMPMent  DMPLdis  DNeiSens  DNeiPhys  DNeiMent  DNeiLdis 

DSchSens  DSchPhys  DSchMent  DSchLdis  DBosSens  DBosPhys  DBosMent  DBosLdis 

DGrpSens  DGrpPhys  DGrpMent  DGrpLdis  DMarSens  DMarPhys  DMarMent  DMarLdis  

DNegShp  DNegFr  FlexAsk

(ORDINAL).

 

/*Secure.

VARIABLE LEVEL

    SecurEmp

((ORDINAL).

 

/*Successful.

VARIABLE LEVEL

GvOKHlth GvOKOld GvOKSec GvOKCrm GvOKUnmp GvOKEnv

(ORDINAL).

 

/*Support-oppose.

VARIABLE LEVEL

HomsBult SchPrv SchVol SchPar2 HospPrv HospVol CarePrv CareVol Vouch

(ORDINAL).

 

/*Willing – Unwilling.

VARIABLE LEVEL

TaxHlth PRENVIR TAXENVIR CUTENVIR

(ORDINAL).

 

(continued on page 94 ff as Private Eye would say)

 

The process will be repeated for missing values, but positive values of 8,9 98,99, 998 ,999 to be treated as missing need to be recoded to negative -8, -9, -98, -99, -998, -999 to be captured by Lo thru -1 declared as missing for almost all variables.  The new negative values will also need ADD VALUE LABELS.

 

The file is in reverse year order and I’m keeping the variables in the order in which they (last) appear.  At a later stage it might be worth re-arranging them by separate topic, plus demographics.  Separate files for separate topics can also be created by extracting selected variables from the “mother” file.

 

As I say, you need to get UKDS to give me permission to send you the Dropbox link to cumulative file: it could be some time before there is a fully correct version and UKDS will probably need even more time to process and upload it.  I’m sure Natcen will be only too pleased to agree.

 

Regards

 

John

 

John F Hall (Mr)

[Retired academic survey researcher]

 

Email:   [hidden email] 

Website: www.surveyresearch.weebly.com

SPSS start page:  www.surveyresearch.weebly.com/1-survey-analysis-workshop

 

 

 

 

 

 

From: Sarah Cohen [mailto:[hidden email]]
Sent: 19 April 2016 11:01
To: '[hidden email]' <[hidden email]>
Subject: Quantitative methods - cumulative file for BSA 1983 - 2014

 

Dear John,

 

I am currently working on research around attitudes towards inequality and redistribution in the UK, alongside Dr Gidon Cohen from Durham University. We have been using the BSA survey and have been busy recoding, hence we were very interested in your recent project. We can appreciate the time and effort to do this. I know perhaps presumptuous but we wondered whether we might ask to have access to your data? Given your expertise it would be very useful to draw on your data to aid our research and add to the robustness of our analysis.

 

If you have any questions about the nature of our research or anything else please let me know.

 

Kind regards

 

Sarah

 

Dr Sarah Cohen

Senior Lecturer

Department of Social Sciences

Northumbria University

Newcastle City Campus

Ellison Place

Newcastle upon Tyne

NE1 8ST

 

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Bruce Weaver]

John, why do you Cc: this mailing list when responding to private e-mails?  Do you ask permission before doing so?  Not everyone appreciates having their everyday e-mail address being posted to the web.  When you feel your reply to a private message contains information beneficial to members of this list, you should remove identifying information before posting, IMO--unless you have permission from your correspondent.  YMMV.

Cheers,
Bruce

John F Hall wrote

Sarah
 
Nice to find someone interested in my work.  
 
Unfortunately I'm not at liberty to distribute my cumulative file unless you
have authorisation from UKDS.  Natcen need to approve the deposit of the
cumulative file, but since I've used files downloaded from UKDS, I suspect
it will be quicker to go via them or, since you are a registered user, ask
them if I can send you the file direct via Dropbox.  I have already sent
them the link to download the beta1 version.
 
I'm currently working on a beta2 version of the "mother" file using a table
generated with syntax supplied by Jon Peck (ex-IBM-SPSS, now retired).  

--- snip ---

[John F Hall]

Bruce

Mea culpa.  I thought Sarah's message was to the QM list which is open; it
was addressed to me, but the subject was Quantitative methods - cumulative
file for BSA 1983 - 2014.  Put it down to fatigue after several days working
on the file and replying in the middle of SPSS sessions.

Hopefully SPSS listers will appreciate the bulk of the message.

John

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Bruce Weaver
Sent: 19 April 2016 17:14
To: [hidden email]
Subject: Re: Quantitative methods - cumulative file for BSA 1983 - 2014

John, Why do you Cc: this mailing list when responding to private e-mails?
Do you ask permission before doing so?  Not everyone appreciates having
their everyday e-mail address being posted to the web.  When you feel your
reply to a private message contains information beneficial to members of
this list, you should remove identifying information before posting,
IMO--unless you have permission from your correspondent.  YMMV.

Cheers,
Bruce

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD