Europeans’ Perception about their Present Household Income
Application of Apriori Algorithm for Association Rule Mining - Unsupervised Learning Homework
I Putu Agastya Harta Pratama
2025-02-04
Introduction
Association Rules learning method are conducted on the data obtained from the European Social Survey, in particular the ESS round 11 in 2023. This particular survey measures Social inequalities in health, Gender in contemporary Europe. This task has set hincfel or “Feeling about household’s income nowadays” as the rule consequent.
Library and Data Import
#install.packages("arulesViz")
library(dplyr)
library(arules)
library(moments)
library(arulesViz)data <- read.csv("ESS11/ESS11.csv")
head(data,5)## name essround edition proddate idno cntry dweight pspwght
## 1 ESS11e02 11 2 20.11.2024 50014 AT 1.1851145 0.3928906
## 2 ESS11e02 11 2 20.11.2024 50030 AT 0.6098981 0.3251533
## 3 ESS11e02 11 2 20.11.2024 50057 AT 1.3923296 4.0000234
## 4 ESS11e02 11 2 20.11.2024 50106 AT 0.5560615 0.1762276
## 5 ESS11e02 11 2 20.11.2024 50145 AT 0.7227953 1.0609399
## pweight anweight nwspol netusoft netustm ppltrst pplfair pplhlp polintr
## 1 0.3309145 0.13001321 90 5 180 5 5 5 1
## 2 0.3309145 0.10759795 90 5 570 10 0 1 2
## 3 0.3309145 1.32366590 30 5 30 6 9 8 2
## 4 0.3309145 0.05831629 15 1 6666 6 6 6 3
## 5 0.3309145 0.35108042 60 5 120 6 3 8 2
## psppsgva actrolga psppipla cptppola trstprl trstlgl trstplc trstplt trstprt
## 1 4 5 4 5 6 9 10 5 5
## 2 3 2 3 2 6 6 4 1 0
## 3 4 4 4 3 7 5 8 4 4
## 4 2 2 2 3 5 6 9 3 3
## 5 3 1 4 3 6 8 8 5 5
## trstep trstun vote prtvtdat prtvtebe prtvtchr prtvtccy prtvtffi prtvtffr
## 1 5 5 1 1 NA NA NA NA NA
## 2 5 5 1 5 NA NA NA NA NA
## 3 7 5 1 5 NA NA NA NA NA
## 4 4 4 2 66 NA NA NA NA NA
## 5 6 8 1 5 NA NA NA NA NA
## prtvgde1 prtvgde2 prtvtegr prtvthhu prtvteis prtvteie prtvteit prtvclt1
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## prtvclt2 prtvclt3 prtvtinl prtvtcno prtvtfpl prtvtept prtvtbrs prtvtesk
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## prtvtgsi prtvtges prtvtdse prtvthch prtvtdgb contplt donprty badge sgnptit
## 1 NA NA NA NA NA 2 2 2 2
## 2 NA NA NA NA NA 2 2 1 1
## 3 NA NA NA NA NA 1 1 2 1
## 4 NA NA NA NA NA 2 2 2 2
## 5 NA NA NA NA NA 2 2 2 1
## pbldmna bctprd pstplonl volunfp clsprty prtcleat prtclebe prtclbhr prtclccy
## 1 2 2 2 2 1 1 NA NA NA
## 2 1 1 2 1 1 5 NA NA NA
## 3 1 1 1 1 1 5 NA NA NA
## 4 2 2 2 2 2 66 NA NA NA
## 5 2 2 2 2 1 5 NA NA NA
## prtclgfi prtclgfr prtclgde prtclegr prtclihu prtcleis prtclfie prtclfit
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## prtclclt prtclhnl prtclcno prtcljpl prtclgpt prtclbrs prtclesk prtclgsi
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## prtclhes prtcldse prtclhch prtcldgb prtdgcl lrscale stflife stfeco stfgov
## 1 NA NA NA NA 2 5 8 6 4
## 2 NA NA NA NA 2 0 9 2 5
## 3 NA NA NA NA 2 3 10 6 5
## 4 NA NA NA NA 6 5 7 4 4
## 5 NA NA NA NA 2 2 9 6 7
## stfdem stfedu stfhlth gincdif freehms hmsfmlsh hmsacld euftf lrnobed loylead
## 1 6 8 8 2 2 4 3 6 4 4
## 2 7 10 10 1 1 5 1 9 2 5
## 3 6 5 8 1 1 5 1 7 5 3
## 4 6 5 2 1 2 4 3 5 2 9
## 5 8 9 9 2 2 4 2 10 1 2
## imsmetn imdfetn impcntr imbgeco imueclt imwbcnt happy sclmeet inprdsc sclact
## 1 2 2 3 7 3 5 8 4 1 3
## 2 2 2 2 6 5 9 9 7 4 4
## 3 1 1 1 9 9 8 9 4 4 3
## 4 2 2 2 6 6 5 7 6 3 3
## 5 1 1 2 10 10 10 9 5 4 3
## crmvct aesfdrk health hlthhmp atchctr atcherp rlgblg rlgdnm rlgdnbat rlgdnacy
## 1 2 2 3 3 10 5 1 1 1 NA
## 2 1 3 2 2 8 8 2 66 6666 NA
## 3 2 3 1 3 9 7 2 66 6666 NA
## 4 2 3 3 3 10 8 1 3 10 NA
## 5 2 1 2 3 10 10 1 1 1 NA
## rlgdnafi rlgdnade rlgdnagr rlgdnhu rlgdnais rlgdnie rlgdnlt rlgdnanl rlgdnno
## 1 NA NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA NA
## rlgdnapl rlgdnapt rlgdnrs rlgdnask rlgdnase rlgdnach rlgdngb rlgblge rlgdnme
## 1 NA NA NA NA NA NA NA 6 66
## 2 NA NA NA NA NA NA NA 2 66
## 3 NA NA NA NA NA NA NA 1 1
## 4 NA NA NA NA NA NA NA 6 66
## 5 NA NA NA NA NA NA NA 6 66
## rlgdebat rlgdeacy rlgdeafi rlgdeade rlgdeagr rlgdehu rlgdeais rlgdeie rlgdelt
## 1 6666 NA NA NA NA NA NA NA NA
## 2 6666 NA NA NA NA NA NA NA NA
## 3 1 NA NA NA NA NA NA NA NA
## 4 6666 NA NA NA NA NA NA NA NA
## 5 6666 NA NA NA NA NA NA NA NA
## rlgdeanl rlgdeno rlgdeapl rlgdeapt rlgders rlgdeask rlgdease rlgdeach rlgdegb
## 1 NA NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA NA
## rlgdgr rlgatnd pray dscrgrp dscrrce dscrntn dscrrlg dscrlng dscretn dscrage
## 1 5 6 5 2 0 0 0 0 0 0
## 2 0 7 6 2 0 0 0 0 0 0
## 3 8 5 3 1 0 0 0 0 0 0
## 4 6 6 3 2 0 0 0 0 0 0
## 5 1 7 7 2 0 0 0 0 0 0
## dscrgnd dscrsex dscrdsb dscroth dscrdk dscrref dscrnap dscrna ctzcntr brncntr
## 1 0 0 0 0 0 0 1 0 1 1
## 2 0 0 0 0 0 0 1 0 1 1
## 3 1 0 0 0 0 0 0 0 1 1
## 4 0 0 0 0 0 0 1 0 2 2
## 5 0 0 0 0 0 0 1 0 1 1
## cntbrthd livecnta lnghom1 lnghom2 feethngr facntr fbrncntc mocntr mbrncntc
## 1 6666 6666 GER 000 1 1 6666 1 6666
## 2 6666 6666 GER ENG 8 1 6666 1 6666
## 3 6666 6666 GER ENG 1 1 6666 1 6666
## 4 RO 1989 GER RUM 2 2 RO 2 RO
## 5 6666 6666 GER 000 1 1 6666 1 6666
## ccnthum ccrdprs wrclmch admrclc testjc34 testjc35 testjc36 testjc37 testjc38
## 1 4 4 4 2 66 66 66 2 2
## 2 3 10 5 3 66 66 66 6 6
## 3 4 8 5 2 66 66 66 3 4
## 4 4 6 4 2 66 66 66 4 4
## 5 5 10 4 3 66 66 66 6 6
## testjc39 testjc40 testjc41 testjc42 vteurmmb vteubcmb ctrlife etfruit eatveg
## 1 3 6 6 6 1 NA 8 3 3
## 2 6 3 3 2 55 NA 8 1 1
## 3 3 6 6 6 1 NA 9 4 3
## 4 4 6 6 6 55 NA 8 2 2
## 5 6 4 2 1 1 NA 9 3 3
## dosprt cgtsmok alcfreq alcwkdy alcwknd icgndra alcbnge height weighta dshltgp
## 1 3 4 3 20 22 1 5 178 90 1
## 2 5 5 3 70 64 2 2 168 74 1
## 3 3 1 4 72 72 2 5 180 95 1
## 4 3 6 7 6666 6666 6 6 167 70 1
## 5 3 1 2 24 48 2 4 168 67 0
## dshltms dshltnt dshltref dshltdk dshltna medtrun medtrnp medtrnt medtroc
## 1 1 0 0 0 0 2 0 0 0
## 2 0 0 0 0 0 2 0 0 0
## 3 1 0 0 0 0 2 0 0 0
## 4 1 0 0 0 0 2 0 0 0
## 5 1 0 0 0 0 2 0 0 0
## medtrnl medtrwl medtrnaa medtroth medtrnap medtrref medtrdk medtrna medtrnu
## 1 0 0 0 0 1 0 0 0 1
## 2 0 0 0 0 1 0 0 0 2
## 3 0 0 0 0 1 0 0 0 1
## 4 0 0 0 0 1 0 0 0 1
## 5 0 0 0 0 1 0 0 0 1
## hlpfmly hlpfmhr trhltacu trhltacp trhltcm trhltch trhltos trhltho trhltht
## 1 2 66 0 0 0 0 0 0 0
## 2 1 2 0 0 0 0 0 0 0
## 3 1 1 0 0 1 0 0 0 0
## 4 2 66 0 0 0 0 0 0 0
## 5 2 66 0 0 0 0 0 1 1
## trhlthy trhltmt trhltpt trhltre trhltsh trhltnt trhltref trhltdk trhltna
## 1 0 0 0 0 0 1 0 0 0
## 2 0 0 0 0 0 1 0 0 0
## 3 0 0 0 0 0 0 0 0 0
## 4 0 0 0 0 0 1 0 0 0
## 5 0 0 0 0 0 0 0 0 0
## fltdpr flteeff slprl wrhpp fltlnl enjlf fltsd cldgng hltprhc hltprhb hltprbp
## 1 1 1 1 3 1 3 1 1 0 1 0
## 2 2 2 3 3 3 4 2 2 1 0 0
## 3 2 2 3 3 1 3 1 2 0 0 0
## 4 2 2 3 2 2 2 2 2 0 1 0
## 5 1 1 1 3 1 3 1 1 0 0 0
## hltpral hltprbn hltprpa hltprpf hltprsd hltprsc hltprsh hltprdi hltprnt
## 1 0 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0 0
## 3 0 0 0 1 0 0 0 0 0
## 4 0 1 0 0 0 0 0 1 0
## 5 0 0 0 0 0 0 0 0 1
## hltprref hltprdk hltprna hltphhc hltphhb hltphbp hltphal hltphbn hltphpa
## 1 0 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0 0
## 3 0 0 0 0 0 0 0 0 1
## 4 0 0 0 0 0 0 0 1 0
## 5 0 0 0 0 0 0 0 0 0
## hltphpf hltphsd hltphsc hltphsh hltphdi hltphnt hltphnap hltphref hltphdk
## 1 0 0 0 0 0 1 0 0 0
## 2 0 0 0 0 0 0 0 1 0
## 3 1 0 0 0 0 0 0 0 0
## 4 0 0 0 0 1 0 0 0 0
## 5 0 0 0 0 0 0 1 0 0
## hltphna hltprca cancfre cnfpplh fnsdfml jbexpvi jbexpti jbexpml jbexpmc
## 1 0 3 6 4 5 0 0 0 0
## 2 0 3 6 2 4 0 0 0 0
## 3 0 3 6 4 2 0 0 0 0
## 4 0 3 6 4 3 0 1 1 0
## 5 0 3 6 5 5 0 0 0 0
## jbexpnt jbexpnap jbexpref jbexpdk jbexpna jbexevl jbexevh jbexevc jbexera
## 1 0 1 0 0 0 0 0 0 0
## 2 1 0 0 0 0 0 0 0 0
## 3 1 0 0 0 0 0 0 0 0
## 4 0 0 0 0 0 0 0 0 0
## 5 1 0 0 0 0 0 0 0 0
## jbexecp jbexebs jbexent jbexenap jbexeref jbexedk jbexena nobingnd likrisk
## 1 0 0 0 1 0 0 0 1 2
## 2 0 0 1 0 0 0 0 2 0
## 3 0 0 1 0 0 0 0 2 0
## 4 1 1 0 0 0 0 0 2 0
## 5 0 0 1 0 0 0 0 2 1
## liklead sothnds actcomp mascfel femifel impbemw trmedmw trwrkmw trplcmw
## 1 4 4 5 4 0 3 3 3 3
## 2 0 6 6 3 6 3 3 2 3
## 3 3 6 6 0 6 5 3 3 3
## 4 1 6 6 0 6 5 3 1 3
## 5 0 6 6 0 5 5 3 3 3
## trmdcnt trwkcnt trplcnt eqwrkbg eqpolbg eqmgmbg eqpaybg eqparep eqparlv
## 1 3 3 3 6 6 6 6 4 4
## 2 1 1 8 6 5 5 5 2 4
## 3 1 1 1 5 6 6 6 2 8
## 4 3 1 2 5 5 5 6 1 8
## 5 3 1 3 6 6 6 6 1 5
## freinsw fineqpy wsekpwr weasoff wlespdm wexashr wprtbym wbrgwrm hhmmb gndr
## 1 4 3 2 3 2 2 2 2 2 1
## 2 1 2 3 3 4 2 2 3 1 2
## 3 2 1 2 2 4 2 4 2 3 2
## 4 4 1 3 3 4 2 3 2 1 2
## 5 1 1 2 2 4 1 3 3 2 1
## gndr2 gndr3 gndr4 gndr5 gndr6 gndr7 gndr8 gndr9 gndr10 gndr11 gndr12 yrbrn
## 1 2 6 6 6 6 6 NA NA NA NA NA 1958
## 2 6 6 6 6 6 6 NA NA NA NA NA 2002
## 3 1 2 6 6 6 6 NA NA NA NA NA 1970
## 4 6 6 6 6 6 6 NA NA NA NA NA 1945
## 5 1 6 6 6 6 6 NA NA NA NA NA 1959
## agea yrbrn2 yrbrn3 yrbrn4 yrbrn5 yrbrn6 yrbrn7 yrbrn8 yrbrn9 yrbrn10 yrbrn11
## 1 65 1957 6666 6666 6666 6666 6666 NA NA NA NA
## 2 21 6666 6666 6666 6666 6666 6666 NA NA NA NA
## 3 53 1970 2001 6666 6666 6666 6666 NA NA NA NA
## 4 78 6666 6666 6666 6666 6666 6666 NA NA NA NA
## 5 64 1962 6666 6666 6666 6666 6666 NA NA NA NA
## yrbrn12 rshipa2 rshipa3 rshipa4 rshipa5 rshipa6 rshipa7 rshipa8 rshipa9
## 1 NA 1 66 66 66 66 66 NA NA
## 2 NA 66 66 66 66 66 66 NA NA
## 3 NA 1 2 66 66 66 66 NA NA
## 4 NA 66 66 66 66 66 66 NA NA
## 5 NA 1 66 66 66 66 66 NA NA
## rshipa10 rshipa11 rshipa12 rshpsts rshpsgb lvgptnea dvrcdeva marsts marstgb
## 1 NA NA NA 1 NA 1 2 66 NA
## 2 NA NA NA 66 NA 2 2 6 NA
## 3 NA NA NA 1 NA 1 2 66 NA
## 4 NA NA NA 66 NA 1 1 4 NA
## 5 NA NA NA 1 NA 1 2 66 NA
## maritalb chldhhe domicil paccmoro paccdwlr pacclift paccnbsh paccocrw
## 1 1 1 3 0 0 0 0 0
## 2 6 2 1 1 0 0 0 0
## 3 1 6 3 0 0 0 0 0
## 4 4 1 1 0 0 0 0 0
## 5 1 1 4 0 0 0 0 0
## paccxhoc paccnois paccinro paccnt paccref paccdk paccna edulvlb eisced
## 1 0 0 0 1 0 0 0 322 3
## 2 0 0 0 0 0 0 0 423 5
## 3 0 0 0 1 0 0 0 610 6
## 4 0 0 0 1 0 0 0 422 5
## 5 0 0 0 1 0 0 0 322 3
## edlveat edlvebe edlvehr edlvgcy edlvdfi edlvdfr edudde1 educde2 edlvegr
## 1 6 NA NA NA NA NA NA NA NA
## 2 10 NA NA NA NA NA NA NA NA
## 3 12 NA NA NA NA NA NA NA NA
## 4 9 NA NA NA NA NA NA NA NA
## 5 6 NA NA NA NA NA NA NA NA
## edlvdahu edlvdis edlvdie edlvfit edlvdlt edlvenl edlveno edlvipl edlvept
## 1 NA NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA NA
## edlvdrs edlvdsk edlvesi edlvies edlvdse edlvdch educgb1 edubgb2 edagegb
## 1 NA NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA NA
## eduyrs pdwrk edctn uempla uempli dsbld rtrd cmsrv hswrk dngoth dngref dngdk
## 1 12 0 0 0 0 0 1 0 0 0 0 0
## 2 14 0 1 0 0 0 0 0 0 0 0 0
## 3 16 1 0 0 0 0 0 0 0 0 0 0
## 4 14 0 0 0 0 0 1 0 0 0 0 0
## 5 12 0 0 0 0 0 1 0 0 0 0 0
## dngna mainact mnactic crpdwk pdjobev pdjobyr emplrel emplno wrkctra estsz
## 1 0 66 6 2 2 6666 6 66666 6 6
## 2 0 66 2 2 1 2023 1 66666 2 2
## 3 0 66 1 6 6 6666 1 66666 1 1
## 4 0 66 6 2 1 2005 1 66666 1 5
## 5 0 66 6 2 1 2021 1 66666 1 2
## jbspv njbspv wkdcorga iorgact wkhct wkhtot nacer2 tporgwk isco08 wrkac6m
## 1 6 66666 66 66 666 666 666 66 66666 6
## 2 2 66666 0 0 32 32 14 4 5249 2
## 3 1 4 10 9 25 25 88 6 2635 2
## 4 2 66666 5 0 39 39 87 2 2221 2
## 5 2 66666 7 7 40 35 47 4 5223 2
## uemp3m uemp12m uemp5yr mbtru hincsrca hinctnta hincfel edulvlpb eiscedp
## 1 2 6 6 3 4 6 1 322 3
## 2 2 6 6 3 8 1 2 6666 66
## 3 2 6 6 3 1 5 1 520 5
## 4 2 6 6 3 4 2 2 6666 66
## 5 2 6 6 2 4 77 2 322 3
## edlvpfat edlvpebe edlvpehr edlvpgcy edlvpdfi edlvpdfr edupdde1 edupcde2
## 1 6 NA NA NA NA NA NA NA
## 2 6666 NA NA NA NA NA NA NA
## 3 11 NA NA NA NA NA NA NA
## 4 6666 NA NA NA NA NA NA NA
## 5 6 NA NA NA NA NA NA NA
## edlvpegr edlvpdahu edlvpdis edlvpdie edlvpfit edlvpdlt edlvpenl edlvpeno
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## edlvphpl edlvpept edlvpdrs edlvpdsk edlvpesi edlvphes edlvpdse edlvpdch
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## edupcgb1 edupbgb2 edagepgb pdwrkp edctnp uemplap uemplip dsbldp rtrdp cmsrvp
## 1 NA NA NA 0 0 0 0 0 1 0
## 2 NA NA NA 0 0 0 0 0 0 0
## 3 NA NA NA 1 0 0 0 0 0 0
## 4 NA NA NA 0 0 0 0 0 0 0
## 5 NA NA NA 0 0 0 0 0 1 0
## hswrkp dngothp dngdkp dngnapp dngrefp dngnap mnactp crpdwkp isco08p emprelp
## 1 0 0 0 0 0 0 66 2 66666 6
## 2 0 0 0 1 0 0 66 6 66666 6
## 3 0 0 0 0 0 0 66 6 2635 1
## 4 0 0 0 1 0 0 66 6 66666 6
## 5 0 0 0 0 0 0 66 2 66666 6
## wkhtotp edulvlfb eiscedf edlvfeat edlvfebe edlvfehr edlvfgcy edlvfdfi
## 1 666 322 3 7 NA NA NA NA
## 2 666 322 3 6 NA NA NA NA
## 3 1 322 3 6 NA NA NA NA
## 4 666 322 3 6 NA NA NA NA
## 5 666 322 3 6 NA NA NA NA
## edlvfdfr edufcde1 edufbde2 edlvfegr edlvfdahu edlvfdis edlvfdie edlvffit
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## edlvfdlt edlvfenl edlvfeno edlvfgpl edlvfept edlvfdrs edlvfdsk edlvfesi
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## edlvfges edlvfdse edlvfdch edufcgb1 edufbgb2 edagefgb emprf14 occf14b
## 1 NA NA NA NA NA NA 2 9
## 2 NA NA NA NA NA NA 2 9
## 3 NA NA NA NA NA NA 1 6
## 4 NA NA NA NA NA NA 1 5
## 5 NA NA NA NA NA NA 1 4
## edulvlmb eiscedm edlvmeat edlvmebe edlvmehr edlvmgcy edlvmdfi edlvmdfr
## 1 212 2 3 NA NA NA NA NA
## 2 322 3 6 NA NA NA NA NA
## 3 322 3 7 NA NA NA NA NA
## 4 212 2 3 NA NA NA NA NA
## 5 212 2 3 NA NA NA NA NA
## edumcde1 edumbde2 edlvmegr edlvmdahu edlvmdis edlvmdie edlvmfit edlvmdlt
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## edlvmenl edlvmeno edlvmgpl edlvmept edlvmdrs edlvmdsk edlvmesi edlvmges
## 1 NA NA NA NA NA NA NA NA
## 2 NA NA NA NA NA NA NA NA
## 3 NA NA NA NA NA NA NA NA
## 4 NA NA NA NA NA NA NA NA
## 5 NA NA NA NA NA NA NA NA
## edlvmdse edlvmdch edumcgb1 edumbgb2 edagemgb emprm14 occm14b atncrse anctrya1
## 1 NA NA NA NA NA 2 9 2 11010
## 2 NA NA NA NA NA 2 9 2 11010
## 3 NA NA NA NA NA 3 66 1 11010
## 4 NA NA NA NA NA 1 4 2 14090
## 5 NA NA NA NA NA 1 7 2 11010
## anctrya2 regunit region ipcrtiva impricha ipeqopta ipshabta impsafea impdiffa
## 1 555555 2 AT31 3 5 2 2 2 4
## 2 555555 2 AT22 2 4 2 4 4 4
## 3 11070 2 AT33 1 4 1 3 2 4
## 4 11010 2 AT31 3 4 2 3 1 4
## 5 11018 2 AT32 3 4 2 2 2 4
## ipfrulea ipudrsta ipmodsta ipgdtima impfreea iphlppla ipsucesa ipstrgva
## 1 2 2 2 2 2 2 3 2
## 2 4 2 2 2 2 1 4 2
## 3 3 1 3 3 2 1 3 2
## 4 3 3 2 3 2 2 3 3
## 5 5 2 3 3 2 2 2 2
## ipadvnta ipbhprpa iprspota iplylfra impenva imptrada impfuna testji1 testji2
## 1 5 2 2 2 2 3 3 6 6
## 2 4 2 4 1 1 4 2 6 6
## 3 4 3 3 1 1 3 2 6 6
## 4 4 3 3 2 2 2 3 6 6
## 5 4 2 2 2 2 2 2 6 6
## testji3 testji4 testji5 testji6 testji7 testji8 testji9 respc19a symtc19
## 1 6 3 3 2 66 66 66 1 2
## 2 6 6 6 6 10 2 2 1 2
## 3 6 8 8 8 66 66 66 1 2
## 4 6 1 1 1 66 66 66 1 2
## 5 6 6 6 6 8 4 3 3 6
## symtnc19 vacc19 recon inwds ainws
## 1 6 1 2 2023-11-12 14:49:50 2023-11-12 14:49:50
## 2 6 1 1 2023-10-18 09:56:32 2023-10-18 09:56:32
## 3 6 1 1 2023-09-30 13:22:49 2023-09-30 13:22:49
## 4 6 1 2 2023-06-30 14:31:46 2023-06-30 14:31:46
## 5 6 1 2 2023-07-11 10:11:00 2023-07-11 10:11:00
## ainwe binwe cinwe
## 1 2023-11-12 14:50:23 2023-11-12 15:01:14 2023-11-12 15:03:07
## 2 2023-10-18 09:58:32 2023-10-18 10:07:21 2023-10-18 10:11:42
## 3 2023-09-30 13:24:24 2023-09-30 13:34:23 2023-09-30 13:41:50
## 4 2023-06-30 14:32:40 2023-06-30 14:42:30 2023-06-30 14:50:31
## 5 2023-07-11 10:13:36 2023-07-11 10:18:33 2023-07-11 10:23:56
## dinwe einwe finwe
## 1 2023-11-12 15:08:28 2023-11-12 15:14:58 2023-11-12 15:18:41
## 2 2023-10-18 10:19:27 2023-10-18 10:25:47 2023-10-18 10:32:34
## 3 2023-09-30 13:50:25 2023-09-30 13:56:10 2023-09-30 14:03:15
## 4 2023-06-30 14:53:56 2023-06-30 14:57:00 2023-06-30 15:04:03
## 5 2023-07-11 10:43:23 2023-07-11 10:56:16 2023-07-11 11:03:31
## hinwe iinwe kinwe rinwe
## 1 2023-11-12 15:20:19 2023-11-12 15:20:37 2023-11-12 15:20:46
## 2 2023-10-18 10:36:39 2023-10-18 10:37:18 2023-10-18 10:37:49
## 3 2023-09-30 14:05:12 2023-09-30 14:05:27 2023-09-30 14:05:41
## 4 2023-06-30 15:05:22 2023-06-30 15:05:41 2023-06-30 15:06:02
## 5 2023-07-11 11:07:49 2023-07-11 11:08:12 2023-07-11 11:08:26
## inwde jinws jinwe inwtm mode domain
## 1 2023-11-12 15:26:55 2023-11-12 15:21:28 2023-11-12 15:26:55 30 1 2
## 2 2023-10-18 10:44:18 2023-10-18 10:42:22 2023-10-18 10:44:18 40 1 1
## 3 2023-09-30 14:13:33 2023-09-30 14:08:31 2023-09-30 14:13:33 42 1 2
## 4 2023-06-30 15:11:21 2023-06-30 15:08:05 2023-06-30 15:11:21 34 1 1
## 5 2023-07-11 11:14:03 2023-07-11 11:10:02 2023-07-11 11:14:03 57 1 2
## prob stratum psu
## 1 0.0005786479 107 317
## 2 0.0011243912 69 128
## 3 0.0004925300 18 418
## 4 0.0012332522 101 295
## 5 0.0009487666 115 344Data Preprocessing
Remove unecessary variables
remove_admin_var <- c("recon", "inwds", "ainws", "ainwe", "binwe", "cinwe", "dinwe",
"einwe", "finwe", "hinwe", "iinwe", "kinwe", "rinwe", "inwde",
"jinws", "jinwe", "inwtm")
remove_stats_var <- c("domain", "prob", "stratum","psu","name","essround","edition","proddate","idno")
remove_covid19 <- c("respc19a", "symtc19", "symtnc19", "vacc19")
data <- data %>% select(-all_of(remove_admin_var))
data <- data %>% select(-all_of(remove_stats_var))
data <- data %>% select(-all_of(remove_covid19))Features Selection
I selected these variables to capture a comprehensive view of financial perception, combining demographic, economic, political, and subjective factors. My goal is to understand how personal characteristics, employment conditions, social trust, and psychological well-being influence financial stability across different income levels.
1. Socio-Demographics - gndr, agea,
marsts: Individual characteristics affecting financial
perception.
- cntry, region, cntbrthd: Geographic and migration
effects on economic well-being.
- hhmmb, chldhhe, domicil: Household structure impacts
financial security.
2. Socio-Economics - eisced, edulvlb,
edulvlpb: Education levels influence job opportunities and
income.
- emplrel, emprelp, wrkctra: Employment stability
determines financial security.
- fnsdfml, hinctnta, hincfel, hincsrca: Direct measures
of income and financial well-being.
- uemp12m, uemp5yr, pdjobev, pdwrkp, isco08, isco08p, tporgwk,
wkhtotp: Work history, job type, and work hours impact
financial resilience.
3. Socio-Political - netusoft:
Media exposure shapes financial perspectives.
- ppltrst, pplfair, pplhlp: Social trust influences
financial behavior.
- polintr, psppsgva, actrolga, psppipla, vote, lrscale:
Political engagement links to financial stability.
- trstprl, trstlgl, trstplt, trstprt, trstep:
Institutional trust may impact economic security.
- gincdif, stfdem, freehms, hmsfmlsh, hmsacld, euftf, imbgeco,
imueclt, imwbcnt: Perceptions of economy, housing, and
inclusivity relate to financial well-being.
4. Subjectiveness and Human Values - happy,
enjlf: Life satisfaction influences financial perception.
- ctrlife, fltdpr, flteeff, fltlnl, fltsd:
Psychological factors affect economic decision-making.
- dscrgrp, dscrrce, dscrntn, dscrrlg, dscretn, dscrgnd,
dscrsex: Perceived discrimination may impact financial
opportunities.
- impricha, ipeqopta, impdiffa, impsafea, ipfrulea,
ipudrsta: Personal values shape financial and social
behaviors.
data_selected <- data %>%
select(
# 1. Socio-Demographics - Age, Gender, Marital Status, Country, Country of Birth, Household Size, Ever had Child?, Domicile size
gndr, agea, marsts, cntry, region, cntbrthd, hhmmb, chldhhe, domicil,
# 2. Socio-Economics: Education Level, Employment Status, Occupation-Related, Income-Related
eisced, edulvlb, emplrel, fnsdfml, uemp12m, uemp5yr, emprelp, pdjobev, edulvlpb, wrkctra, isco08, isco08p, tporgwk, pdwrkp, hinctnta, hincfel, wkhtotp, hincsrca,
# 3. Socio-Political: Media Exposure, Social Trust, Political Trust, Democracy Perception, Perception for Minorities
netusoft, ppltrst, pplfair, pplhlp, polintr, psppsgva, actrolga, psppipla, trstprl, trstlgl, trstplt, trstprt, trstep, vote, lrscale, gincdif, stfdem, freehms, hmsfmlsh, hmsacld, euftf, imbgeco, imueclt, imwbcnt,
# 4 Subjectiveness and Human Values
happy, dscrgrp, dscrrce, dscrntn, dscrrlg, dscretn, dscrgnd, dscrsex, ctrlife, fltdpr, flteeff, fltlnl, enjlf, fltsd, impricha, ipeqopta, impdiffa, impsafea, ipfrulea, ipudrsta
)Check Missing Values Before Conversion
Because there are so many NAs (missing values) I’m looking to see which variables has the highest percentage of missing value. Later, I’m removing values that are not correlated
na_counts_before <- colSums(is.na(data_selected))
na_columns_before <- na_counts_before[na_counts_before > 0]
na_percentage_before <- (na_columns_before / nrow(data_selected)) * 100
na_summary_before <- data.frame(Variable = names(na_columns_before),
NA_Count = na_columns_before,
NA_Percentage_From_Total = na_percentage_before)
na_summary_before## Variable NA_Count NA_Percentage_From_Total
## cntbrthd cntbrthd 3 0.007470864From this we can see that only one variable (cntbrthd) has missing values and it’s very low so we are good to proceed.
Checking Unique Values in each variables
unique_values <- lapply(data_selected, unique)
unique_values_df <- data.frame(
Variable = names(unique_values),
Unique_Values = sapply(unique_values, function(x) paste(head(x, 10), collapse = ", ")),
Unique_Count = sapply(data_selected, function(x) length(unique(x)))
)
unique_values_df## Variable Unique_Values
## gndr gndr 1, 2
## agea agea 65, 21, 53, 78, 64, 59, 77, 69, 52, 75
## marsts marsts 66, 6, 4, 5, 1, 2, 77, 88, 99, 3
## cntry cntry AT, BE, CH, CY, DE, ES, FI, FR, GB, GR
## region region AT31, AT22, AT33, AT32, AT12, AT11, AT13, AT34, AT21, BE24
## cntbrthd cntbrthd 6666, RO, CZ, IR, MA, MK, DE, XK, RS, SI
## hhmmb hhmmb 2, 1, 3, 4, 5, 7, 0, 6, 8, 77
## chldhhe chldhhe 1, 2, 6, 7, 9, 8
## domicil domicil 3, 1, 4, 5, 2, 8, 7, 9
## eisced eisced 3, 5, 6, 4, 2, 7, 55, 1, 77, 88
## edulvlb edulvlb 322, 423, 610, 422, 313, 212, 720, 213, 520, 710
## emplrel emplrel 6, 1, 2, 3, 7, 8, 9
## fnsdfml fnsdfml 5, 4, 2, 3, 1, 7, 8, 9
## uemp12m uemp12m 6, 2, 1, 7, 8, 9
## uemp5yr uemp5yr 6, 2, 1, 7, 9, 8
## emprelp emprelp 6, 1, 2, 3, 9, 8, 7
## pdjobev pdjobev 2, 1, 6, 7, 9, 8
## edulvlpb edulvlpb 322, 6666, 520, 720, 313, 610, 113, 212, 423, 422
## wrkctra wrkctra 6, 2, 1, 3, 8, 9, 7
## isco08 isco08 66666, 5249, 2635, 2221, 5223, 1112, 8219, 2330, 4226, 9629
## isco08p isco08p 66666, 2635, 3321, 3343, 4110, 7412, 2149, 2341, 5246, 5223
## tporgwk tporgwk 66, 4, 6, 2, 1, 5, 3, 88, 77, 99
## pdwrkp pdwrkp 0, 1
## hinctnta hinctnta 6, 1, 5, 2, 77, 9, 3, 10, 8, 4
## hincfel hincfel 1, 2, 4, 3, 8, 7, 9
## wkhtotp wkhtotp 666, 1, 88, 77, 45, 40, 39, 50, 32, 38
## hincsrca hincsrca 4, 8, 1, 5, 2, 3, 6, 7, 77, 88
## netusoft netusoft 5, 1, 2, 3, 4, 7, 8, 9
## ppltrst ppltrst 5, 10, 6, 8, 7, 3, 4, 2, 1, 9
## pplfair pplfair 5, 0, 9, 6, 3, 8, 7, 4, 2, 10
## pplhlp pplhlp 5, 1, 8, 6, 4, 7, 3, 9, 2, 10
## polintr polintr 1, 2, 3, 4, 8, 9, 7
## psppsgva psppsgva 4, 3, 2, 1, 7, 8, 5, 9
## actrolga actrolga 5, 2, 4, 1, 3, 8, 7, 9
## psppipla psppipla 4, 3, 2, 1, 5, 8, 7, 9
## trstprl trstprl 6, 7, 5, 3, 9, 8, 2, 4, 0, 77
## trstlgl trstlgl 9, 6, 5, 8, 7, 4, 2, 88, 3, 10
## trstplt trstplt 5, 1, 4, 3, 7, 8, 6, 9, 0, 2
## trstprt trstprt 5, 0, 4, 3, 6, 2, 8, 9, 7, 1
## trstep trstep 5, 7, 4, 6, 8, 2, 3, 88, 0, 1
## vote vote 1, 2, 3, 7, 8, 9
## lrscale lrscale 5, 0, 3, 2, 4, 77, 6, 88, 9, 7
## gincdif gincdif 2, 1, 3, 5, 8, 7, 4, 9
## stfdem stfdem 6, 7, 8, 3, 9, 0, 4, 1, 2, 5
## freehms freehms 2, 1, 3, 4, 5, 7, 8, 9
## hmsfmlsh hmsfmlsh 4, 5, 3, 2, 7, 8, 1, 9
## hmsacld hmsacld 3, 1, 2, 4, 8, 5, 7, 9
## euftf euftf 6, 9, 7, 5, 10, 3, 1, 4, 8, 88
## imbgeco imbgeco 7, 6, 9, 10, 5, 88, 8, 3, 1, 4
## imueclt imueclt 3, 5, 9, 6, 10, 8, 0, 7, 4, 1
## imwbcnt imwbcnt 5, 9, 8, 10, 6, 2, 7, 0, 3, 4
## happy happy 8, 9, 7, 5, 10, 6, 3, 4, 0, 1
## dscrgrp dscrgrp 2, 1, 7, 8, 9
## dscrrce dscrrce 0, 1
## dscrntn dscrntn 0, 1
## dscrrlg dscrrlg 0, 1
## dscretn dscretn 0, 1
## dscrgnd dscrgnd 0, 1
## dscrsex dscrsex 0, 1
## ctrlife ctrlife 8, 9, 6, 10, 5, 7, 4, 2, 1, 3
## fltdpr fltdpr 1, 2, 3, 4, 8, 7, 9
## flteeff flteeff 1, 2, 3, 8, 4, 7, 9
## fltlnl fltlnl 1, 3, 2, 4, 8, 7, 9
## enjlf enjlf 3, 4, 2, 1, 8, 7, 9
## fltsd fltsd 1, 2, 3, 4, 8, 7, 9
## impricha impricha 5, 4, 2, 6, 3, 88, 77, 1, 66, 99
## ipeqopta ipeqopta 2, 1, 3, 88, 4, 5, 6, 77, 66, 99
## impdiffa impdiffa 4, 2, 3, 1, 88, 6, 5, 77, 66, 99
## impsafea impsafea 2, 4, 1, 3, 88, 77, 5, 6, 66, 99
## ipfrulea ipfrulea 2, 4, 3, 5, 1, 88, 6, 77, 66, 99
## ipudrsta ipudrsta 2, 1, 3, 4, 88, 5, 6, 77, 66, 99
## Unique_Count
## gndr 2
## agea 77
## marsts 10
## cntry 24
## region 264
## cntbrthd 196
## hhmmb 16
## chldhhe 6
## domicil 8
## eisced 11
## edulvlb 31
## emplrel 7
## fnsdfml 8
## uemp12m 6
## uemp5yr 6
## emprelp 7
## pdjobev 6
## edulvlpb 32
## wrkctra 7
## isco08 568
## isco08p 523
## tporgwk 10
## pdwrkp 2
## hinctnta 13
## hincfel 7
## wkhtotp 95
## hincsrca 11
## netusoft 8
## ppltrst 14
## pplfair 14
## pplhlp 14
## polintr 7
## psppsgva 8
## actrolga 8
## psppipla 8
## trstprl 14
## trstlgl 14
## trstplt 14
## trstprt 14
## trstep 14
## vote 6
## lrscale 14
## gincdif 8
## stfdem 14
## freehms 8
## hmsfmlsh 8
## hmsacld 8
## euftf 14
## imbgeco 14
## imueclt 14
## imwbcnt 14
## happy 14
## dscrgrp 5
## dscrrce 2
## dscrntn 2
## dscrrlg 2
## dscretn 2
## dscrgnd 2
## dscrsex 2
## ctrlife 14
## fltdpr 7
## flteeff 7
## fltlnl 7
## enjlf 7
## fltsd 7
## impricha 10
## ipeqopta 10
## impdiffa 10
## impsafea 10
## ipfrulea 10
## ipudrsta 10This table above indicate that there are some variables that have obscenely high unique value counts within them. Upon doing rechecks, I discover that some of them are designed to have multiple levels within, although some e.g. age (agea), number of people living regularly as member of household (hhmmb), hours normally worked a week in main job overtime included, partner (wkhtotp) are inherently numerical. Thus, I have to do further preprocessing on those features, before converting them into transactional format.
Inspection Consequent Variable
str(data_selected$hincfel)## int [1:40156] 1 2 1 2 2 1 2 1 2 2 ...From here we can see that hincfel or “Feeling about household’s income nowadays” is categorized into 7 levels. For the purpose of this project, I am keeping only 4
- 1 Living comfortably on present income
- 2 Coping on present income
- 3 Difficult on present income
- 4 Very difficult on present income
Next I am converting all of the variables to factor before converting to transaction.
vars_exclude <- c("agea", "hhmmb", "wkhtotp")
data_selected <- data_selected %>%
mutate(across(-all_of(vars_exclude), as.factor))str(data_selected$agea)## int [1:40156] 65 21 53 78 64 59 77 69 52 75 ...str(data_selected$hhmmb)## int [1:40156] 2 1 3 1 2 2 1 2 3 2 ...str(data_selected$wkhtotp)## int [1:40156] 666 666 1 666 666 1 666 666 1 666 ...Encoding Variables
Firstly, there are some invalid values within this dataset, so I have to filter them out.
# Handle invalid values for hhmmb (Household members)
data_selected$hhmmb[data_selected$hhmmb %in% c(77, 88, 99)] <- NA
# Handle invalid values for agea (Age)
data_selected$agea[data_selected$agea > 120 | data_selected$agea == 999] <- NA # Age should be realistic
# Handle invalid values for wkhtotp (Weekly working hours)
data_selected$wkhtotp[data_selected$wkhtotp > 168 | data_selected$wkhtotp %in% c(666, 777, 888, 999)] <- NAThen I proceed with the encoding
data_selected <- data_selected %>%
mutate(agea = case_when(
agea < 18 ~ "Under 18",
agea >= 18 & agea <= 29 ~ "18-29",
agea >= 30 & agea <= 39 ~ "30-39",
agea >= 40 & agea <= 49 ~ "40-49",
agea >= 50 & agea <= 59 ~ "50-59",
agea >= 60 & agea <= 69 ~ "60-69",
agea >= 70 ~ "70+",
TRUE ~ as.character(NA) # Keeps NA for missing or invalid values
))
data_selected <- data_selected %>%
mutate(wkhtotp = case_when(
wkhtotp == 0 ~ "Unemployed / No Work",
wkhtotp > 0 & wkhtotp <= 10 ~ "1-10 hrs (Very Part-Time)",
wkhtotp > 10 & wkhtotp <= 20 ~ "11-20 hrs (Part-Time)",
wkhtotp > 20 & wkhtotp <= 35 ~ "21-35 hrs (Reduced Full-Time)",
wkhtotp > 35 & wkhtotp <= 45 ~ "36-45 hrs (Standard Full-Time)",
wkhtotp > 45 & wkhtotp <= 60 ~ "46-60 hrs (Overtime Heavy)",
wkhtotp > 60 & wkhtotp <= 80 ~ "61-80 hrs (Extreme Workload)",
wkhtotp > 80 & wkhtotp <= 168 ~ "80+ hrs (Workaholic)",
TRUE ~ as.character(NA) # Keeps NA for missing or invalid values
))
data_selected <- data_selected %>%
mutate(hhmmb = case_when(
hhmmb == 1 ~ "Living Alone",
hhmmb == 2 ~ "Couple / Two People",
hhmmb >= 3 & hhmmb <= 4 ~ "Small Household (3-4)",
hhmmb >= 5 & hhmmb <= 6 ~ "Medium Household (5-6)",
hhmmb >= 7 ~ "Large Household (7+)",
TRUE ~ as.character(NA) # Keeps NA for missing or invalid values
))
data_selected$agea <- as.factor(data_selected$agea)
data_selected$wkhtotp <- as.factor(data_selected$wkhtotp)
data_selected$hhmmb <- as.factor(data_selected$hhmmb)Read as Transaction
data_transaction <- as(data_selected, "transactions")summary(data_transaction)## transactions as itemMatrix in sparse format with
## 40156 rows (elements/itemsets/transactions) and
## 2197 columns (items) and a density of 0.03200523
##
## most frequent items:
## dscretn=0 dscrsex=0 dscrrce=0 dscrrlg=0 dscrntn=0 (Other)
## 39767 39767 39675 39660 39575 2625145
##
## element (itemset/transaction) length distribution:
## sizes
## 68 69 70 71
## 38 369 26635 13114
##
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 68.00 70.00 70.00 70.32 71.00 71.00
##
## includes extended item information - examples:
## labels variables levels
## 1 gndr=1 gndr 1
## 2 gndr=2 gndr 2
## 3 agea=18-29 agea 18-29
##
## includes extended transaction information - examples:
## transactionID
## 1 1
## 2 2
## 3 3inspect(data_transaction[1:5]) ## items transactionID
## [1] {gndr=1,
## agea=60-69,
## marsts=66,
## cntry=AT,
## region=AT31,
## cntbrthd=6666,
## hhmmb=Couple / Two People,
## chldhhe=1,
## domicil=3,
## eisced=3,
## edulvlb=322,
## emplrel=6,
## fnsdfml=5,
## uemp12m=6,
## uemp5yr=6,
## emprelp=6,
## pdjobev=2,
## edulvlpb=322,
## wrkctra=6,
## isco08=66666,
## isco08p=66666,
## tporgwk=66,
## pdwrkp=0,
## hinctnta=6,
## hincfel=1,
## hincsrca=4,
## netusoft=5,
## ppltrst=5,
## pplfair=5,
## pplhlp=5,
## polintr=1,
## psppsgva=4,
## actrolga=5,
## psppipla=4,
## trstprl=6,
## trstlgl=9,
## trstplt=5,
## trstprt=5,
## trstep=5,
## vote=1,
## lrscale=5,
## gincdif=2,
## stfdem=6,
## freehms=2,
## hmsfmlsh=4,
## hmsacld=3,
## euftf=6,
## imbgeco=7,
## imueclt=3,
## imwbcnt=5,
## happy=8,
## dscrgrp=2,
## dscrrce=0,
## dscrntn=0,
## dscrrlg=0,
## dscretn=0,
## dscrgnd=0,
## dscrsex=0,
## ctrlife=8,
## fltdpr=1,
## flteeff=1,
## fltlnl=1,
## enjlf=3,
## fltsd=1,
## impricha=5,
## ipeqopta=2,
## impdiffa=4,
## impsafea=2,
## ipfrulea=2,
## ipudrsta=2} 1
## [2] {gndr=2,
## agea=18-29,
## marsts=6,
## cntry=AT,
## region=AT22,
## cntbrthd=6666,
## hhmmb=Living Alone,
## chldhhe=2,
## domicil=1,
## eisced=5,
## edulvlb=423,
## emplrel=1,
## fnsdfml=4,
## uemp12m=6,
## uemp5yr=6,
## emprelp=6,
## pdjobev=1,
## edulvlpb=6666,
## wrkctra=2,
## isco08=5249,
## isco08p=66666,
## tporgwk=4,
## pdwrkp=0,
## hinctnta=1,
## hincfel=2,
## hincsrca=8,
## netusoft=5,
## ppltrst=10,
## pplfair=0,
## pplhlp=1,
## polintr=2,
## psppsgva=3,
## actrolga=2,
## psppipla=3,
## trstprl=6,
## trstlgl=6,
## trstplt=1,
## trstprt=0,
## trstep=5,
## vote=1,
## lrscale=0,
## gincdif=1,
## stfdem=7,
## freehms=1,
## hmsfmlsh=5,
## hmsacld=1,
## euftf=9,
## imbgeco=6,
## imueclt=5,
## imwbcnt=9,
## happy=9,
## dscrgrp=2,
## dscrrce=0,
## dscrntn=0,
## dscrrlg=0,
## dscretn=0,
## dscrgnd=0,
## dscrsex=0,
## ctrlife=8,
## fltdpr=2,
## flteeff=2,
## fltlnl=3,
## enjlf=4,
## fltsd=2,
## impricha=4,
## ipeqopta=2,
## impdiffa=4,
## impsafea=4,
## ipfrulea=4,
## ipudrsta=2} 2
## [3] {gndr=2,
## agea=50-59,
## marsts=66,
## cntry=AT,
## region=AT33,
## cntbrthd=6666,
## hhmmb=Small Household (3-4),
## chldhhe=6,
## domicil=3,
## eisced=6,
## edulvlb=610,
## emplrel=1,
## fnsdfml=2,
## uemp12m=6,
## uemp5yr=6,
## emprelp=1,
## pdjobev=6,
## edulvlpb=520,
## wrkctra=1,
## isco08=2635,
## isco08p=2635,
## tporgwk=6,
## pdwrkp=1,
## hinctnta=5,
## hincfel=1,
## wkhtotp=1-10 hrs (Very Part-Time),
## hincsrca=1,
## netusoft=5,
## ppltrst=6,
## pplfair=9,
## pplhlp=8,
## polintr=2,
## psppsgva=4,
## actrolga=4,
## psppipla=4,
## trstprl=7,
## trstlgl=5,
## trstplt=4,
## trstprt=4,
## trstep=7,
## vote=1,
## lrscale=3,
## gincdif=1,
## stfdem=6,
## freehms=1,
## hmsfmlsh=5,
## hmsacld=1,
## euftf=7,
## imbgeco=9,
## imueclt=9,
## imwbcnt=8,
## happy=9,
## dscrgrp=1,
## dscrrce=0,
## dscrntn=0,
## dscrrlg=0,
## dscretn=0,
## dscrgnd=1,
## dscrsex=0,
## ctrlife=9,
## fltdpr=2,
## flteeff=2,
## fltlnl=1,
## enjlf=3,
## fltsd=1,
## impricha=4,
## ipeqopta=1,
## impdiffa=4,
## impsafea=2,
## ipfrulea=3,
## ipudrsta=1} 3
## [4] {gndr=2,
## agea=70+,
## marsts=4,
## cntry=AT,
## region=AT31,
## cntbrthd=RO,
## hhmmb=Living Alone,
## chldhhe=1,
## domicil=1,
## eisced=5,
## edulvlb=422,
## emplrel=1,
## fnsdfml=3,
## uemp12m=6,
## uemp5yr=6,
## emprelp=6,
## pdjobev=1,
## edulvlpb=6666,
## wrkctra=1,
## isco08=2221,
## isco08p=66666,
## tporgwk=2,
## pdwrkp=0,
## hinctnta=2,
## hincfel=2,
## hincsrca=4,
## netusoft=1,
## ppltrst=6,
## pplfair=6,
## pplhlp=6,
## polintr=3,
## psppsgva=2,
## actrolga=2,
## psppipla=2,
## trstprl=5,
## trstlgl=6,
## trstplt=3,
## trstprt=3,
## trstep=4,
## vote=2,
## lrscale=5,
## gincdif=1,
## stfdem=6,
## freehms=2,
## hmsfmlsh=4,
## hmsacld=3,
## euftf=5,
## imbgeco=6,
## imueclt=6,
## imwbcnt=5,
## happy=7,
## dscrgrp=2,
## dscrrce=0,
## dscrntn=0,
## dscrrlg=0,
## dscretn=0,
## dscrgnd=0,
## dscrsex=0,
## ctrlife=8,
## fltdpr=2,
## flteeff=2,
## fltlnl=2,
## enjlf=2,
## fltsd=2,
## impricha=4,
## ipeqopta=2,
## impdiffa=4,
## impsafea=1,
## ipfrulea=3,
## ipudrsta=3} 4
## [5] {gndr=1,
## agea=60-69,
## marsts=66,
## cntry=AT,
## region=AT32,
## cntbrthd=6666,
## hhmmb=Couple / Two People,
## chldhhe=1,
## domicil=4,
## eisced=3,
## edulvlb=322,
## emplrel=1,
## fnsdfml=5,
## uemp12m=6,
## uemp5yr=6,
## emprelp=6,
## pdjobev=1,
## edulvlpb=322,
## wrkctra=1,
## isco08=5223,
## isco08p=66666,
## tporgwk=4,
## pdwrkp=0,
## hinctnta=77,
## hincfel=2,
## hincsrca=4,
## netusoft=5,
## ppltrst=6,
## pplfair=3,
## pplhlp=8,
## polintr=2,
## psppsgva=3,
## actrolga=1,
## psppipla=4,
## trstprl=6,
## trstlgl=8,
## trstplt=5,
## trstprt=5,
## trstep=6,
## vote=1,
## lrscale=2,
## gincdif=2,
## stfdem=8,
## freehms=2,
## hmsfmlsh=4,
## hmsacld=2,
## euftf=10,
## imbgeco=10,
## imueclt=10,
## imwbcnt=10,
## happy=9,
## dscrgrp=2,
## dscrrce=0,
## dscrntn=0,
## dscrrlg=0,
## dscretn=0,
## dscrgnd=0,
## dscrsex=0,
## ctrlife=9,
## fltdpr=1,
## flteeff=1,
## fltlnl=1,
## enjlf=3,
## fltsd=1,
## impricha=4,
## ipeqopta=2,
## impdiffa=4,
## impsafea=2,
## ipfrulea=5,
## ipudrsta=2} 5Top Frequent Items
itemFrequencyPlot(data_transaction, topN=20, type="absolute", main="Top 25 Most Frequent Items")
I notice that lack of reported discrimination
(dscreth=0, dscrsex=0, dscrrce=0,
etc.), foreign-born status
(cntbrthd=6666), and long-term
unemployment (uemp5yr=6, uemp12m=6)
are highly prevalent. Additionally, job instability
(isco08p=66666, emprelp=6,
wrkctra=1), political participation
(vote=1), and low digital engagement
(netusoft=5) frequently appear. These attributes
likely play a crucial role in shaping association rules, influencing how
different socio-economic factors relate to financial perception.
Rule Discovery - Apriori algorithm
Apriori algorith is used to mine the association rules between each transactions. As mentioned, the rule consequent has been set up as such.
First Rule - Comfortable with their Present Income
Feeling about household’s income nowadays: “Living comfortably on present income” Threshold that I have set for this particular consequent are as follows:
- Support (
supp = 0.01): Ensures that rules appear in at least 1% of transactions, filtering out extremely rare patterns while keeping significant ones.
- Confidence (
conf = 0.6): Requires rules to be at least 60% reliable, meaning that when antecedents occur, the consequent (hincfel=1) follows 60% of the time.
rules_comfortable <- apriori(data_transaction,
parameter = list(supp = 0.01, conf = 0.6, minlen = 2),
appearance = list(rhs = "hincfel=1", default = "lhs"),
control = list(verbose = FALSE))
inspect(sort(rules_comfortable, by = "confidence", decreasing = TRUE)[1:10])## lhs rhs support confidence
## [1] {eisced=7, hinctnta=10, hmsacld=1} => {hincfel=1} 0.01075804 0.8888889
## [2] {hinctnta=10, freehms=1, happy=9} => {hincfel=1} 0.01148023 0.8848369
## [3] {hinctnta=10, hmsfmlsh=5, happy=9} => {hincfel=1} 0.01220241 0.8781362
## [4] {hinctnta=10, hmsacld=1, flteeff=1} => {hincfel=1} 0.01434406 0.8767123
## [5] {fnsdfml=5, hinctnta=10, hmsacld=1} => {hincfel=1} 0.01267557 0.8745704
## [6] {tporgwk=4, hinctnta=10, hmsacld=1} => {hincfel=1} 0.01250125 0.8730435
## [7] {hinctnta=10, freehms=1, ctrlife=8} => {hincfel=1} 0.01177906 0.8726937
## [8] {eisced=7, fnsdfml=5, hinctnta=10} => {hincfel=1} 0.01155494 0.8689139
## [9] {hinctnta=10, hmsfmlsh=5, ctrlife=8} => {hincfel=1} 0.01230202 0.8681898
## [10] {hinctnta=10, pplfair=7, hmsfmlsh=5} => {hincfel=1} 0.01040940 0.8672199
## coverage lift count
## [1] 0.01210280 2.494529 432
## [2] 0.01297440 2.483158 461
## [3] 0.01389581 2.464354 490
## [4] 0.01636119 2.460358 576
## [5] 0.01449348 2.454347 509
## [6] 0.01431916 2.450062 502
## [7] 0.01349736 2.449080 473
## [8] 0.01329814 2.438473 464
## [9] 0.01416974 2.436441 494
## [10] 0.01200319 2.433719 418Based on the result obtained from the first consequent condition
hincfel=1, which states that people feel, they are living
comfortably with their present income, reveal that high income
(hinctnta=10) is the strongest predictor,
appearing in every rule, indicating that financial stability is
essential for comfort. Higher education (eisced=7),
financial security (fnsdfml=5), and union membership
(tporgwk=4) also play significant roles,
suggesting that access to quality education and job stability contribute
to financial well-being. Psychological and social factors,
including happiness (happy=9), a sense of control over life
(ctrlife=8), and fairness perception
(pplfair=7), further influence financial comfort,
highlighting that financial well-being is not solely income-dependent
but also linked to subjective perceptions. Additionally, free
housing (freehms=1) and progressive social views
(hmsacld=1, support for LGBTQ+ adoption rights)
emerge as relevant factors, potentially reflecting the correlation
between financial security and social attitudes. With high confidence
(86.7%–88.9%) and lift values (~2.4–2.5), these rules confirm that
individuals with higher income, education, financial stability,
socially-liberal stance, strong social trust, and access to stable
housing are significantly more likely to feel financially
comfortable.
plot(rules_comfortable[1:10], method = "graph", engine = "igraph")
The graph-based visualization for rules_comfortable
(hincfel=1) highlights that high income
(hinctnta=10, hinctnta=9), higher education
(edulvlpb=710, edulvlpb=720), and residing in
wealthier European countries (cntry=NL,
cntry=SE, cntry=IS) are key
contributors to financial comfort, while the absence of reported racial
(dscrrce=0) and national discrimination
(dscrntn=0) suggests that economic security in these groups
is more strongly linked to education and income rather than experiences
of social exclusion.
plot(rules_comfortable, method = "grouped", control = list(k = 20))
The grouped matrix visualization for rules_comfortable
(hincfel=1) highlights that high income
(hinctnta=10), economic security (imbgeco=8),
and strong educational background (edulvlpb=710,
edulvlpb=6666) are dominant predictors of
financial comfort, while political interest
(polintr=3), fairness perception (ipfrulea=2),
and perceived economic safety (impsafea=5) also
contribute to financial well-being, with larger support values
indicating that these patterns are widely observed, and higher lift
values suggesting that these factors strongly differentiate those living
comfortably from the rest.
Second Rule - Coping with their present Income
Feeling about household’s income nowadays: “Coping on present income” Threshold that I have set for this particular consequent are as follows:
- Support (
supp = 0.01): Ensures that rules appear in at least 1% of transactions, filtering out extremely rare patterns while keeping significant ones.
- Confidence (
conf = 0.6): Requires rules to be at least 60% reliable, meaning that when antecedents occur, the consequent (hincfel=2) follows 60% of the time.
rules_coping <- apriori(data_transaction,
parameter = list(supp = 0.01, conf = 0.6, minlen = 2),
appearance = list(rhs = "hincfel=2", default = "lhs"),
control = list(verbose = FALSE))
inspect(sort(rules_coping, by = "confidence", decreasing = TRUE)[1:10])## lhs rhs support confidence
## [1] {marsts=66, cntry=PL, wrkctra=1} => {hincfel=2} 0.01108178 0.7685665
## [2] {marsts=66, cntry=PL, emplrel=1} => {hincfel=2} 0.01247634 0.7648855
## [3] {cntry=PL, wrkctra=1, flteeff=1} => {hincfel=2} 0.01038450 0.7637363
## [4] {cntry=PL, emplrel=1, fltsd=1} => {hincfel=2} 0.01200319 0.7626582
## [5] {cntry=PL, emplrel=1, ipeqopta=2} => {hincfel=2} 0.01008567 0.7612782
## [6] {cntry=PL, wrkctra=1, hincsrca=1} => {hincfel=2} 0.01055882 0.7584973
## [7] {cntry=PL, chldhhe=6, dscrgrp=2} => {hincfel=2} 0.01001096 0.7584906
## [8] {cntry=PL, wrkctra=1, fltlnl=1} => {hincfel=2} 0.01282498 0.7573529
## [9] {cntry=PL, vote=1, impsafea=2} => {hincfel=2} 0.01023508 0.7555147
## [10] {marsts=66, cntry=PL, fltdpr=1} => {hincfel=2} 0.01018528 0.7546125
## coverage lift count
## [1] 0.01441877 1.704360 445
## [2] 0.01631139 1.696197 501
## [3] 0.01359697 1.693649 417
## [4] 0.01573862 1.691258 482
## [5] 0.01324833 1.688198 405
## [6] 0.01392071 1.682031 424
## [7] 0.01319853 1.682016 402
## [8] 0.01693396 1.679493 515
## [9] 0.01354717 1.675417 411
## [10] 0.01349736 1.673416 409Based on the result obtained from the second consequent condition
hincfel=2, which states that people feel, they are coping
with their present income, reveal that stable employment
(wrkctra=1, unlimited contract; emplrel=1,
employee status employed) is a key factor in financial
stability, as individuals with steady jobs in Poland
(cntry=PL) are significantly more likely to be
coping rather than struggling. Psychological resilience
(flteeff=1, rarely feeling overwhelmed;
fltdpr=1, low depression levels) also plays a
major role, suggesting that those with a positive mental state are more
likely to manage financially. Additionally, family structure
(marsts=66, not in a traditional marriage) and household
income sources (hincsrca=1, salary-based income)
further contribute to economic stability, reinforcing the importance of
both personal and financial security. Social and institutional trust
also influence financial perception, as individuals who value
safety (impsafea=2) and fairness
(pplfair=7) are more likely to report coping
rather than experiencing financial hardship. The Polish
respondents in these rules exhibit patterns where
employment security, psychological well-being, and trust in
institutions are critical factors for those managing on their
present income. With confidence levels between
75.4%–76.8% and lift values around 1.67–1.70,
these rules highlight the importance of job stability and
socio-economic perceptions in financial resilience within
Poland.
plot(rules_coping[1:10], method = "graph", engine = "igraph")
The graph-based visualization for rules_coping
(hincfel=2) shows that living in Poland
(cntry=PL), Hungary (cntry=HU), and Lithuania
(cntry=LT), having a small household
(hhmmb=Small Household (3-4)), and access to free housing
(freehms=2) are associated with financial coping,
while having an unlimited work contract (wrkctra=1)
and moderate political interest (polintr=3)
indicate a stable but cautious financial position, with
perceived lack of political influence (psppipla=2)
and fairness in society (ipfrulea=2) suggesting
that while individuals are managing, they may feel disconnected from
broader economic and political decision-making.
plot(rules_coping, method = "grouped", control = list(k = 20))
The grouped matrix visualization for rules_coping
(hincfel=2) highlights that moderate household
income (hinctnta=2), stable but lower-status employment
(emprelp=6, isco08p=66666), and small
household size (hhmmb=Small Household (3-4)) are
key indicators of financial coping, while perceived fairness
(ipfrulea=3), social trust (psppsgva=2), and
life satisfaction (enjlf=3) suggest a moderate but
stable economic outlook, with lower lift values compared to
hincfel=1, indicating that coping individuals
experience more variability in financial stability.
Third Rule - Feeling difficult on their present income
Feeling about household’s income nowadays: “Difficult on present income”
Threshold that I have set for this particular consequent are as follows. Additionally, I lower this because I think financial difficulty is less frequent than financial comfort, so a higher threshold would filter out too many potential patterns.
- Support (
supp = 0.005): Lowered to 0.5% to capture less frequent but meaningful patterns, as financial difficulty (hincfel=3) may be less common in the dataset.
- Confidence (
conf = 0.5): Set at 50% to ensure moderately strong rules while allowing for variability in economic hardship factors.
rules_difficult <- apriori(data_transaction,
parameter = list(supp = 0.005, conf = 0.5, minlen = 2),
appearance = list(rhs = "hincfel=3", default = "lhs"),
control = list(verbose = FALSE))
inspect(sort(rules_difficult, by = "confidence", decreasing = TRUE)[1:10])## lhs rhs support confidence
## [1] {region=EL52, uemp12m=6} => {hincfel=3} 0.005603148 0.5754476
## [2] {region=EL52, uemp5yr=6} => {hincfel=3} 0.005603148 0.5754476
## [3] {cntry=GR, region=EL52, uemp12m=6} => {hincfel=3} 0.005603148 0.5754476
## [4] {cntry=GR, region=EL52, uemp5yr=6} => {hincfel=3} 0.005603148 0.5754476
## [5] {region=EL52, uemp12m=6, uemp5yr=6} => {hincfel=3} 0.005603148 0.5754476
## [6] {region=EL52, uemp12m=6, dscrgnd=0} => {hincfel=3} 0.005603148 0.5754476
## [7] {region=EL52, uemp12m=6, dscrrlg=0} => {hincfel=3} 0.005603148 0.5754476
## [8] {region=EL52, uemp12m=6, dscrrce=0} => {hincfel=3} 0.005603148 0.5754476
## [9] {region=EL52, uemp12m=6, dscretn=0} => {hincfel=3} 0.005603148 0.5754476
## [10] {region=EL52, uemp12m=6, dscrsex=0} => {hincfel=3} 0.005603148 0.5754476
## coverage lift count
## [1] 0.009737026 3.941943 225
## [2] 0.009737026 3.941943 225
## [3] 0.009737026 3.941943 225
## [4] 0.009737026 3.941943 225
## [5] 0.009737026 3.941943 225
## [6] 0.009737026 3.941943 225
## [7] 0.009737026 3.941943 225
## [8] 0.009737026 3.941943 225
## [9] 0.009737026 3.941943 225
## [10] 0.009737026 3.941943 225Based on the result obtained from the third consequent condition
hincfel=3, which states that people feel, they are having
difficulties with their present income, reveal that long-term
unemployment (uemp12m=6, unemployment lasting 12 months or
more) and recent unemployment within the last five years
(uemp5yr=6) are strong indicators of financial
difficulty. Individuals in Greece (cntry=GR) and
specifically in region EL52 (Central Greece
Macedonia) are disproportionately represented, suggesting that
regional economic conditions significantly impact perceived financial
difficulty. Furthermore, discrimination-related factors
(dscrrce=0, dscrrlg=0, dscretn=0,
dscrsex=0, dscrgnd=0) appear in
multiple rules, indicating that individuals who do not perceive
discrimination are still struggling financially, pointing to systemic
economic challenges rather than social exclusion as the primary driver.
With moderate confidence levels (57.5%) but a high lift value of
3.94, these findings suggest that long-term
unemployment, regional economic conditions, and systemic financial
barriers are critical factors for individuals experiencing financial
difficulty in Greece.
plot(rules_difficult[1:10], method = "graph", engine = "igraph")
The graph-based visualization for rules_difficult
(hincfel=3) highlights that long-term unemployment
(uemp12m=6, uemp5yr=6), being foreign-born
(cntbrthd=6666), and residing in Greece
(cntry=GR) and region EL52 are key
predictors of financial difficulty, with low digital engagement
(netusoft=5, infrequent internet use) and perceived
discrimination (dscrgrp=2) further reinforcing
economic strain, while the presence of voting behavior
(vote=1) suggests political participation despite financial
struggles, and the absence of reported gender discrimination
(dscrgnd=0) indicates that structural economic issues
rather than direct social exclusion are driving financial hardship.
plot(rules_difficult, method = "grouped", control = list(k = 20))
The grouped matrix visualization for rules_difficult
(hincfel=3) shows that long-term unemployment
(uemp12m=6, uemp5yr=6), foreign-born status
(cntbrthd=6666), and low occupational status
(isco08p=66666, emprelp=6) are strong
predictors of financial difficulty, while low digital engagement
(netusoft=1), low social trust (dscrgrp=2),
and limited financial security (hinscsrca=4)
further reinforce economic hardship, with higher lift values
(3.5–3.9) indicating that these factors strongly differentiate
financially struggling individuals from others.
Fourth Rule - Struggling with their present income
Feeling about household’s income nowadays: “Very difficult on present income”
Threshold that I have set for this particular consequent are as follows. Additionaly, I think that severe financial hardship is rare in this data, so stricter thresholds would eliminate almost all rules.
- Support (
supp = 0.001): Reduced to 0.1% to capture rare but critical patterns, as extreme financial hardship (hincfel=4) is the least frequent category.
- Confidence (
conf = 0.3): Lowered to 30% to retain weaker but still relevant associations, as financial distress can have diverse contributing factors.
rules_very_difficult <- apriori(data_transaction,
parameter = list(supp = 0.001, conf = 0.3, minlen = 2),
appearance = list(rhs = "hincfel=4", default = "lhs"),
control = list(verbose = FALSE))
inspect(sort(rules_very_difficult, by = "confidence", decreasing = TRUE)[1:10])## lhs rhs support confidence
## [1] {fnsdfml=1, hinctnta=1, psppipla=1} => {hincfel=4} 0.001145532 0.5348837
## [2] {cntry=GR, fnsdfml=2, trstprt=0} => {hincfel=4} 0.001319853 0.5196078
## [3] {fnsdfml=1, polintr=4, dscrgrp=1} => {hincfel=4} 0.001070824 0.5119048
## [4] {cntry=GR, fnsdfml=2, trstprl=0} => {hincfel=4} 0.001120630 0.5113636
## [5] {edulvlpb=6666, hinctnta=1, fltsd=4} => {hincfel=4} 0.001120630 0.5056180
## [6] {uemp12m=1, hincsrca=6, gincdif=1} => {hincfel=4} 0.001543978 0.4960000
## [7] {cntry=GR, fnsdfml=2, trstplt=0} => {hincfel=4} 0.001419464 0.4956522
## [8] {uemp5yr=1, hincsrca=6, gincdif=1} => {hincfel=4} 0.001394561 0.4955752
## [9] {hinctnta=1, fltsd=4} => {hincfel=4} 0.001195338 0.4897959
## [10] {hinctnta=1, dscrgnd=0, fltsd=4} => {hincfel=4} 0.001195338 0.4897959
## coverage lift count
## [1] 0.002141648 15.06227 46
## [2] 0.002540094 14.63210 53
## [3] 0.002091842 14.41518 43
## [4] 0.002191453 14.39994 45
## [5] 0.002216356 14.23815 45
## [6] 0.003112860 13.96730 62
## [7] 0.002863831 13.95751 57
## [8] 0.002814025 13.95534 56
## [9] 0.002440482 13.79260 48
## [10] 0.002440482 13.79260 48Based on the result obtained from the fourth consequent condition
hincfel=4, which states that people feel, they are having
severe difficulties with their present income, reveal that low
income (hinctnta=1, bottom income decile), financial strain
(fnsdfml=1, finding it very difficult to make ends meet),
and lack of political influence (psppipla=1, believing they
have no influence on politics) are key factors strongly
associated with extreme financial hardship. Individuals from
Greece (cntry=GR) frequently appear in the
rules, indicating that regional economic conditions contribute
significantly to financial distress. Additionally, low trust in
political institutions (trstprl=0, no trust in parliament;
trstplt=0, no trust in politicians) is a recurring
theme, suggesting a link between financial struggles and political
disenchantment. Lower education levels
(edulvlpb=6666, incomplete primary education) also
appear, reinforcing the correlation between limited education
and economic hardship. Employment instability is another crucial factor,
as individuals who experienced unemployment in the last 12
months (uemp12m=1) or last five years
(uemp5yr=1) are highly likely to struggle
financially. With confidence values around 49%–53% and
exceptionally high lift values (~13.7–15.0), these findings
highlight that low income, unemployment, lack of education, and
distrust in political institutions are key predictors of severe
financial difficulty, particularly in Greece.
plot(rules_very_difficult[1:10], method = "graph", engine = "igraph")
The graph shows that for rules_very_difficult
(hincfel=4) unstable employment
(emprelp=6, pdwrkp=0), unclassified
occupations (isco08p=66666), and low-income levels
(hinctnta=1) are strong predictors of extreme
financial difficulty, with foreign-born individuals
(cntbrthd=6666) and lack of political trust
(trstprl=0, trstplt=0) further
reinforcing economic vulnerability, while the absence of reported
discrimination (dscrrce=0, dscrrlg=0,
dscretn=0) suggests that systemic financial instability,
rather than social exclusion, is a primary driver of financial
hardship.
plot(rules_very_difficult, method = "grouped", control = list(k = 20))
The grouped matrix visualization for
rules_very_difficult (hincfel=4) highlights
that low financial security (hinctnta=3), low
education levels (eisced=3), and weak occupational
stability (wrkctra=1) are key drivers of severe
financial difficulty, while living alone
(hhmmb=Living Alone), limited savings
(hinscsrca=4), and economic dissatisfaction
(imueclt=3, eutrfr=0) further
contribute to financial struggles, with exceptionally high lift
values (10–14) indicating that these characteristics strongly
distinguish those experiencing extreme financial hardship from the
rest.