SlideShare a Scribd company logo
1 of 5
Exploratory data miningonEmployeeTurnover Viabilityandbenefitson a “lowinvestment” case Lisbon ,  August 2009 Paulo Xavier PRPlathXavier@gmail.com
ExecutiveSummary Data miningprojects , particularlyon HR issues, tend to beseen as being expensive highriskdue to uncertaintyofbenefitsorreturn onlyviableifsupportedbyhighlevel HRMS or “Data warehouse” and applied to verylargeorganizations. Thefollowing case showedthat a lotcanbedonewithnoneofthoserequirements, withimmediatbenefits, altoughrequiring  a goodknowledgeofthesubject (organizationalbehaviour & HR Mgmt) and some knowledgeonthetechnologybehind data miningandanalyticsin general. The Case:  A minimum data collection, onanorganizationofaprox 1500 employees, wasenough to identify some ofthekeyfactorsthat stand behindemployeeturnover, exposingspecificareas for immediateintervention, allowinganestimatedreductiononaprox 50% oftheoccurringandundesiredturnover. Increasedinvestmentsandtimecanobviouslyimproveresults. Data mining on Employee turnover data - Paulo Xavier - August 2009
Data collectedandanalysisprocedure Onlyconsideredexitsduringonespecificyear (with a particularlyhighnumberofexits). Thiswas a yearwithhighlabourmarketactivity . Appartfrominformationof general interest (Age, Seniority, genderetc) itwasmostly data concerningtheemployeesCompensationand Performance thatwasincluded. Final data to betreatedincluded1580 recordsand34variablesfromwhichonlyhalfproved to berelevant. Severalalgoritmsweretriedbutthesimplestapproachprovedthemostreliableandproducedthemostinterestingresults. Trainingofthemodelwasdoneon a sampleof 70% ofthe data andtestingontheremaining 30%. Fromseveralinitialteststhosevariablesthatprovedirrelevantweregraduallyexcluded. The final version, whoseresults are presented, isnottheonethatproved more accurateinpredictingturnover, Howeverit´stheonethatincludesvariablesofrelevance for futureactions/decisionsand for thatreasonconsideredthe final one.  Inthis case, more thenaccuracyinprediction , thecapacity to turnconclusionsinactionabledecisionswasconsideredrelevant.  All final conclusions are translatableinspecificactions/decisionsthatinfluenceturnover.  Impactisestimatedinbeingable to reduceturnoverinaprox 50% versus previoussituation. Data mining on Employee turnover data - Paulo Xavier - August 2009
DecisionTreeandrules Thefollowingdecisiontreeandnext slide withrules are the final resultfromthisexercise. Theyidentifyareaswerecompensationadjustments are necessary to reassurethereductionofturnover. Nonrewardissues are obviouslyalsorelevant to justifyturnover, butinthis case a verylimitedsetofvariableswasenough to obtaintangibleandusefullresults.    Minimumexitprobability = 0 Maximum = 1 Undereachrulenumber, thelikelihoodofexit Ex: Rule127lkelihoodofexit = 0,8148 (for 27 observationsinthetestsample) Data mining on Employee turnover data - Paulo Xavier - August 2009
Rules Noticerulesnr 127, 59 and 58 Top Performers (Assess = A and B ) are identifiedandscoredonriskofexit. Theinclusionof Performance assessmentresults, allowstheidentificationofthose cases werespecialmeasures for adequateprotectionofhigherperformersisneeded.   Rulenumber: 2 [cover=748 (68%)]    range.lev.1.evol>=0.015   Rulenumber: 6 [cover=246 (22%)]    range.lev.1.evol< 0.015 Senior_< 1.2   Rulenumber: 127 [cover=27 (2%)]    range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5    Rng.pos.Level.1< 0.635 Assess=A,B Senior_< 5.955   Rulenumber: 28 [cover=20 (2%)]    range.lev.1.evol< 0.015 Senior_>=1.2    Age>=37.5 Assess=C,D,N/A   Rulenumber: 125 [cover=15 (1%)]    range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5    Rng.pos.Level.1< 0.635 Assess=C,D Senior_< 2.675   Rulenumber: 59 [cover=14 (1%)]    range.lev.1.evol< 0.015 Senior_>=1.2    Age>=37.5 Assess=A,B Senior_< 15.38 Rulenumber: 30 [cover=10 (1%)]    range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5    Rng.pos.Level.1>=0.635   Rulenumber: 58 [cover=8 (1%)]    range.lev.1.evol< 0.015 Senior_>=1.2    Age>=37.5 Assess=A,B Senior_>=15.38   Rulenumber: 126 [cover=7 (1%)]    range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5    Rng.pos.Level.1< 0.635 Assess=A,B Senior_>=5.955   Rulenumber: 124 [cover=7 (1%)]    range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5    Rng.pos.Level.1< 0.635 Assess=C,D Senior_>=2.675 Data mining on Employee turnover data - Paulo Xavier - August 2009

More Related Content

Viewers also liked

Fort St John Goalkeeper Camp
Fort St John Goalkeeper CampFort St John Goalkeeper Camp
Fort St John Goalkeeper Campsheldano
 
Watch Powerpoint Presentation
Watch Powerpoint PresentationWatch Powerpoint Presentation
Watch Powerpoint Presentationhenleycolmedia
 
Jinx Powerpoint Presentation
Jinx Powerpoint PresentationJinx Powerpoint Presentation
Jinx Powerpoint Presentationhenleycolmedia
 
Pictures from JEEcamp
Pictures from JEEcampPictures from JEEcamp
Pictures from JEEcampguestce72c9
 
Esperson Residence-Advanced 30
Esperson Residence-Advanced 30Esperson Residence-Advanced 30
Esperson Residence-Advanced 30tcirata
 
LINGKUNGAN STRATEGIK BALI DR YANI
LINGKUNGAN STRATEGIK BALI   DR YANILINGKUNGAN STRATEGIK BALI   DR YANI
LINGKUNGAN STRATEGIK BALI DR YANIYani Antariksa
 
Penjelasan fgd alki laksda yani tim dan penanggap utama
Penjelasan fgd alki laksda yani tim dan penanggap utamaPenjelasan fgd alki laksda yani tim dan penanggap utama
Penjelasan fgd alki laksda yani tim dan penanggap utamaYani Antariksa
 
Welcome To The World Of Samaritan Help Mission
Welcome To The World Of Samaritan Help MissionWelcome To The World Of Samaritan Help Mission
Welcome To The World Of Samaritan Help MissionMamoon Akhtar
 
Data Mining As Used In Employee Recruitment &
Data Mining As Used In Employee Recruitment &Data Mining As Used In Employee Recruitment &
Data Mining As Used In Employee Recruitment &melodysmithjones
 
IMPLEMENTASI NILAI KBS PANCASILA,
IMPLEMENTASI  NILAI KBS PANCASILA, IMPLEMENTASI  NILAI KBS PANCASILA,
IMPLEMENTASI NILAI KBS PANCASILA, Yani Antariksa
 
Digital Migration SA
Digital Migration SADigital Migration SA
Digital Migration SAguestcb98d54
 
Graduate employability and industry partnership
Graduate employability and industry partnershipGraduate employability and industry partnership
Graduate employability and industry partnershipRéda ALLAL
 
Employee Attrition
Employee AttritionEmployee Attrition
Employee AttritionVinay sattur
 

Viewers also liked (14)

Fort St John Goalkeeper Camp
Fort St John Goalkeeper CampFort St John Goalkeeper Camp
Fort St John Goalkeeper Camp
 
Watch Powerpoint Presentation
Watch Powerpoint PresentationWatch Powerpoint Presentation
Watch Powerpoint Presentation
 
Jinx Powerpoint Presentation
Jinx Powerpoint PresentationJinx Powerpoint Presentation
Jinx Powerpoint Presentation
 
Pictures from JEEcamp
Pictures from JEEcampPictures from JEEcamp
Pictures from JEEcamp
 
Esperson Residence-Advanced 30
Esperson Residence-Advanced 30Esperson Residence-Advanced 30
Esperson Residence-Advanced 30
 
LINGKUNGAN STRATEGIK BALI DR YANI
LINGKUNGAN STRATEGIK BALI   DR YANILINGKUNGAN STRATEGIK BALI   DR YANI
LINGKUNGAN STRATEGIK BALI DR YANI
 
Penjelasan fgd alki laksda yani tim dan penanggap utama
Penjelasan fgd alki laksda yani tim dan penanggap utamaPenjelasan fgd alki laksda yani tim dan penanggap utama
Penjelasan fgd alki laksda yani tim dan penanggap utama
 
Welcome To The World Of Samaritan Help Mission
Welcome To The World Of Samaritan Help MissionWelcome To The World Of Samaritan Help Mission
Welcome To The World Of Samaritan Help Mission
 
Data Mining As Used In Employee Recruitment &
Data Mining As Used In Employee Recruitment &Data Mining As Used In Employee Recruitment &
Data Mining As Used In Employee Recruitment &
 
IMPLEMENTASI NILAI KBS PANCASILA,
IMPLEMENTASI  NILAI KBS PANCASILA, IMPLEMENTASI  NILAI KBS PANCASILA,
IMPLEMENTASI NILAI KBS PANCASILA,
 
Digital Migration SA
Digital Migration SADigital Migration SA
Digital Migration SA
 
Graduate employability and industry partnership
Graduate employability and industry partnershipGraduate employability and industry partnership
Graduate employability and industry partnership
 
Employee Attrition
Employee AttritionEmployee Attrition
Employee Attrition
 
attrition analysis
attrition analysisattrition analysis
attrition analysis
 

Similar to Exploratory Data Mining On Employee Turnover

ClintWorld User Conference 2015 - Stefan Schwarz
ClintWorld User Conference 2015 - Stefan SchwarzClintWorld User Conference 2015 - Stefan Schwarz
ClintWorld User Conference 2015 - Stefan SchwarzStefan Schwarz
 
Implement Data Ware House
Implement Data Ware HouseImplement Data Ware House
Implement Data Ware Housebhuphender
 
Smart Staffing using Regression Analysis Model
Smart Staffing using Regression Analysis ModelSmart Staffing using Regression Analysis Model
Smart Staffing using Regression Analysis ModelAnand Narayanan
 
WhyInvestPROCESSCONTROL
WhyInvestPROCESSCONTROLWhyInvestPROCESSCONTROL
WhyInvestPROCESSCONTROLPierre Latour
 
PARTNERS 2014 - Dr. Stefan Schwarz - Money for Nothing
PARTNERS 2014 - Dr. Stefan Schwarz - Money for NothingPARTNERS 2014 - Dr. Stefan Schwarz - Money for Nothing
PARTNERS 2014 - Dr. Stefan Schwarz - Money for NothingStefan Schwarz
 
pickadatactrsiteprintedition copy
pickadatactrsiteprintedition copypickadatactrsiteprintedition copy
pickadatactrsiteprintedition copyRobyn Weisman
 
Dynamic Resource Allocation Strategy CL 12-16-2014
Dynamic Resource Allocation Strategy CL 12-16-2014Dynamic Resource Allocation Strategy CL 12-16-2014
Dynamic Resource Allocation Strategy CL 12-16-2014Christopher Lietz
 
IRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET Journal
 
Attrition Calculation
Attrition CalculationAttrition Calculation
Attrition Calculationswati18
 
Cybersecurity Incident Management PowerPoint Presentation Slides
Cybersecurity Incident Management PowerPoint Presentation SlidesCybersecurity Incident Management PowerPoint Presentation Slides
Cybersecurity Incident Management PowerPoint Presentation SlidesSlideTeam
 
Business Opportunities, Challenges, Strategies and Execution in Big Data Era ...
Business Opportunities, Challenges, Strategies and Execution in Big Data Era...Business Opportunities, Challenges, Strategies and Execution in Big Data Era...
Business Opportunities, Challenges, Strategies and Execution in Big Data Era ...Craig Chao
 
Transforming Retail Workforce
Transforming Retail WorkforceTransforming Retail Workforce
Transforming Retail WorkforceAnil Kumar
 
A better approach to managing information
A better approach to managing informationA better approach to managing information
A better approach to managing informationMark Albala
 
Cybersecurity Incident Management Powerpoint Presentation Slides
Cybersecurity Incident Management Powerpoint Presentation SlidesCybersecurity Incident Management Powerpoint Presentation Slides
Cybersecurity Incident Management Powerpoint Presentation SlidesSlideTeam
 
TechMD - Backup vs Business Continuity
TechMD - Backup vs Business ContinuityTechMD - Backup vs Business Continuity
TechMD - Backup vs Business ContinuityTechMD
 
Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...
Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...
Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...*instinctools
 
Fahim Karim: Attrition Prevention
Fahim Karim: Attrition PreventionFahim Karim: Attrition Prevention
Fahim Karim: Attrition PreventionEdunomica
 
Pay Me Now or Pay Me A Lot More Later
Pay Me Now or Pay Me A Lot More LaterPay Me Now or Pay Me A Lot More Later
Pay Me Now or Pay Me A Lot More LaterRLE Technologies
 
Combating Fraud: Six Principles for Security
Combating Fraud: Six Principles for Security Combating Fraud: Six Principles for Security
Combating Fraud: Six Principles for Security Strategic Treasurer
 

Similar to Exploratory Data Mining On Employee Turnover (20)

ClintWorld User Conference 2015 - Stefan Schwarz
ClintWorld User Conference 2015 - Stefan SchwarzClintWorld User Conference 2015 - Stefan Schwarz
ClintWorld User Conference 2015 - Stefan Schwarz
 
Implement Data Ware House
Implement Data Ware HouseImplement Data Ware House
Implement Data Ware House
 
Smart Staffing using Regression Analysis Model
Smart Staffing using Regression Analysis ModelSmart Staffing using Regression Analysis Model
Smart Staffing using Regression Analysis Model
 
WhyInvestPROCESSCONTROL
WhyInvestPROCESSCONTROLWhyInvestPROCESSCONTROL
WhyInvestPROCESSCONTROL
 
PARTNERS 2014 - Dr. Stefan Schwarz - Money for Nothing
PARTNERS 2014 - Dr. Stefan Schwarz - Money for NothingPARTNERS 2014 - Dr. Stefan Schwarz - Money for Nothing
PARTNERS 2014 - Dr. Stefan Schwarz - Money for Nothing
 
pickadatactrsiteprintedition copy
pickadatactrsiteprintedition copypickadatactrsiteprintedition copy
pickadatactrsiteprintedition copy
 
Dynamic Resource Allocation Strategy CL 12-16-2014
Dynamic Resource Allocation Strategy CL 12-16-2014Dynamic Resource Allocation Strategy CL 12-16-2014
Dynamic Resource Allocation Strategy CL 12-16-2014
 
IRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data Mining
 
Attrition Calculation
Attrition CalculationAttrition Calculation
Attrition Calculation
 
Cybersecurity Incident Management PowerPoint Presentation Slides
Cybersecurity Incident Management PowerPoint Presentation SlidesCybersecurity Incident Management PowerPoint Presentation Slides
Cybersecurity Incident Management PowerPoint Presentation Slides
 
Di&Tingting_5.0 v
Di&Tingting_5.0 vDi&Tingting_5.0 v
Di&Tingting_5.0 v
 
Business Opportunities, Challenges, Strategies and Execution in Big Data Era ...
Business Opportunities, Challenges, Strategies and Execution in Big Data Era...Business Opportunities, Challenges, Strategies and Execution in Big Data Era...
Business Opportunities, Challenges, Strategies and Execution in Big Data Era ...
 
Transforming Retail Workforce
Transforming Retail WorkforceTransforming Retail Workforce
Transforming Retail Workforce
 
A better approach to managing information
A better approach to managing informationA better approach to managing information
A better approach to managing information
 
Cybersecurity Incident Management Powerpoint Presentation Slides
Cybersecurity Incident Management Powerpoint Presentation SlidesCybersecurity Incident Management Powerpoint Presentation Slides
Cybersecurity Incident Management Powerpoint Presentation Slides
 
TechMD - Backup vs Business Continuity
TechMD - Backup vs Business ContinuityTechMD - Backup vs Business Continuity
TechMD - Backup vs Business Continuity
 
Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...
Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...
Data Integration: Huntflow and PowerBI | Case Study | Software Development Co...
 
Fahim Karim: Attrition Prevention
Fahim Karim: Attrition PreventionFahim Karim: Attrition Prevention
Fahim Karim: Attrition Prevention
 
Pay Me Now or Pay Me A Lot More Later
Pay Me Now or Pay Me A Lot More LaterPay Me Now or Pay Me A Lot More Later
Pay Me Now or Pay Me A Lot More Later
 
Combating Fraud: Six Principles for Security
Combating Fraud: Six Principles for Security Combating Fraud: Six Principles for Security
Combating Fraud: Six Principles for Security
 

Exploratory Data Mining On Employee Turnover

  • 1. Exploratory data miningonEmployeeTurnover Viabilityandbenefitson a “lowinvestment” case Lisbon , August 2009 Paulo Xavier PRPlathXavier@gmail.com
  • 2. ExecutiveSummary Data miningprojects , particularlyon HR issues, tend to beseen as being expensive highriskdue to uncertaintyofbenefitsorreturn onlyviableifsupportedbyhighlevel HRMS or “Data warehouse” and applied to verylargeorganizations. Thefollowing case showedthat a lotcanbedonewithnoneofthoserequirements, withimmediatbenefits, altoughrequiring a goodknowledgeofthesubject (organizationalbehaviour & HR Mgmt) and some knowledgeonthetechnologybehind data miningandanalyticsin general. The Case: A minimum data collection, onanorganizationofaprox 1500 employees, wasenough to identify some ofthekeyfactorsthat stand behindemployeeturnover, exposingspecificareas for immediateintervention, allowinganestimatedreductiononaprox 50% oftheoccurringandundesiredturnover. Increasedinvestmentsandtimecanobviouslyimproveresults. Data mining on Employee turnover data - Paulo Xavier - August 2009
  • 3. Data collectedandanalysisprocedure Onlyconsideredexitsduringonespecificyear (with a particularlyhighnumberofexits). Thiswas a yearwithhighlabourmarketactivity . Appartfrominformationof general interest (Age, Seniority, genderetc) itwasmostly data concerningtheemployeesCompensationand Performance thatwasincluded. Final data to betreatedincluded1580 recordsand34variablesfromwhichonlyhalfproved to berelevant. Severalalgoritmsweretriedbutthesimplestapproachprovedthemostreliableandproducedthemostinterestingresults. Trainingofthemodelwasdoneon a sampleof 70% ofthe data andtestingontheremaining 30%. Fromseveralinitialteststhosevariablesthatprovedirrelevantweregraduallyexcluded. The final version, whoseresults are presented, isnottheonethatproved more accurateinpredictingturnover, Howeverit´stheonethatincludesvariablesofrelevance for futureactions/decisionsand for thatreasonconsideredthe final one. Inthis case, more thenaccuracyinprediction , thecapacity to turnconclusionsinactionabledecisionswasconsideredrelevant. All final conclusions are translatableinspecificactions/decisionsthatinfluenceturnover. Impactisestimatedinbeingable to reduceturnoverinaprox 50% versus previoussituation. Data mining on Employee turnover data - Paulo Xavier - August 2009
  • 4. DecisionTreeandrules Thefollowingdecisiontreeandnext slide withrules are the final resultfromthisexercise. Theyidentifyareaswerecompensationadjustments are necessary to reassurethereductionofturnover. Nonrewardissues are obviouslyalsorelevant to justifyturnover, butinthis case a verylimitedsetofvariableswasenough to obtaintangibleandusefullresults. Minimumexitprobability = 0 Maximum = 1 Undereachrulenumber, thelikelihoodofexit Ex: Rule127lkelihoodofexit = 0,8148 (for 27 observationsinthetestsample) Data mining on Employee turnover data - Paulo Xavier - August 2009
  • 5. Rules Noticerulesnr 127, 59 and 58 Top Performers (Assess = A and B ) are identifiedandscoredonriskofexit. Theinclusionof Performance assessmentresults, allowstheidentificationofthose cases werespecialmeasures for adequateprotectionofhigherperformersisneeded. Rulenumber: 2 [cover=748 (68%)] range.lev.1.evol>=0.015   Rulenumber: 6 [cover=246 (22%)] range.lev.1.evol< 0.015 Senior_< 1.2   Rulenumber: 127 [cover=27 (2%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5 Rng.pos.Level.1< 0.635 Assess=A,B Senior_< 5.955   Rulenumber: 28 [cover=20 (2%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age>=37.5 Assess=C,D,N/A   Rulenumber: 125 [cover=15 (1%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5 Rng.pos.Level.1< 0.635 Assess=C,D Senior_< 2.675   Rulenumber: 59 [cover=14 (1%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age>=37.5 Assess=A,B Senior_< 15.38 Rulenumber: 30 [cover=10 (1%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5 Rng.pos.Level.1>=0.635   Rulenumber: 58 [cover=8 (1%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age>=37.5 Assess=A,B Senior_>=15.38   Rulenumber: 126 [cover=7 (1%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5 Rng.pos.Level.1< 0.635 Assess=A,B Senior_>=5.955   Rulenumber: 124 [cover=7 (1%)] range.lev.1.evol< 0.015 Senior_>=1.2 Age< 37.5 Rng.pos.Level.1< 0.635 Assess=C,D Senior_>=2.675 Data mining on Employee turnover data - Paulo Xavier - August 2009