APPLICATION OF DATAMINING TOOL FOR
CLASSIFICATION OF ORGANIZATIONAL
CHANGE EXPECTATION
Şule ÖZMEN
Serra YURTKORU
Beril SİPAHİ
DATA MINING
Data mining is the nontrivial extraction
of implicit, previously unknown, and
potentially useful information from data.
DIFFERENT GOALS CALL FOR
DIFFERENT TECHNIQUES
DATAMINING TECHNIQUES
Datamining techniques can be either
directed
or
undirected.
DATAMINING TECHNIQUES
Directed
Undirected
Goal is
to predict, estimate,
classify, or characterize
the behavior of some
pre-identified
target
variable
Goal is
to discover structure in
the data set as a whole.
DATAMINING TECHNIQUES
Directed
• Classification
• Estimation
• Prediction
Undirected
• Description &
Visualization
• Association Rule or
Affinity Grouping
• Clustering
CLASSIFICATION
Classification is used to develop a model
that maps a data item into one of several
predefined classes.
DECISION TREE ANALYSIS
Builds classification and regression trees
Starts with pre-identified target variable in
other words dependent variable. This is the
initial node
Initial node is split into two or more child
nodes
Splitting is based on statistical analysis
used by decision tree algorithms
DECISION TREE ANALYSIS
Target Variable
Target Variable
Predictive
Variables
DECISION TREE ALGORITHMS
CHAID (Chi square Automatic Interaction
Detector)
C&RT (Classification and Regression Tree)
QUEST (Quick Unbiased Efficient Statistical
Test)
CHAID Method
CHAID was designed to handle categorical
variables only.
SPSS has extended algorithm to handle
nominal, ordinal and continuous dependent
variables.
Components of CHAID
One or more predictor variables. Predictor
variables can be continuous, ordinal, or
nominal.
One target variable. The target variable can be
nominal, ordinal or continuous.
CHAID Algorithms
A CHAID tree is a decision tree
that is constructed by splitting
subsets of the space into two or
more child (nodes) repeatedly,
beginning with the entire data set.
CHAID Algorithms
To determine the best split at any node,
CHAID merges any allowable pair of
categories of the predictor variable (the set
of allowable pairs is determined by the
type of predictor variable being studied) if
there is no statistically significant
difference within the pair with respect to
the target variable.
CHAID Algorithms
The process is repeated until no nonsignificant pair is found. The resulting set
of categories of the predictor variable is the
best split with respect to that predictor
variable. This process is followed for all
predictor variables. The split that is the
best prediction is selected, and the node is
split. The process repeats recursively until
one of the stopping rules is triggered.
APPLICATION
AIM OF THE RESEARCH
The ability to be both receptive and
responsive
to
change
has
become
important in recent years.
Therefore our aim is to analyze change
patterns that will help managers and
organizations to manage the process of
change more effectively
SAMPLE
Our sample is consisted of 253 subjects
from 7 private Turkish organizations.
The sample is composed of 44 superiors
and 209 subordinates.
INSTRUMENT
Multi Scale Organizational Change
Questionnaire
Organizational change questionnaire is
composed of five scales "Forces of
Change", “Change Strategy", "Means of
Change", “Resistance to Change", and
“Change Expectation" scales
TARGET VARIABLE
Change Expectation
•Employee Development *
•Efficiency
•Organization Structure
•Acquisition & Divestiture
•Alliances
•Restructuring
* means increase in employee self development & individual learning,
increase in employee participation& employee suggestions acceptance
Since organizational change is a process
that takes time, we rather asked if the
employees expected change as a result of
the actions taken within the firm, not
whether the organization has changed or
not. This is also important because if
the employees don’t believe in the
actions taken, they resist and try to block
the change actions.
PREDICTOR VARIABLES
Change Forces
Change Strategy
Means of Change
Resistance to Change
SCALE
P R E D IC T O R S
O rg aniz ation al
B us in es s In pu ts
C H A NG E F OR C E S
C om p etition
L aw s & R egu lations
P r es s ur e G r oups
C H A NG E ST R AT EG Y
M EANS OF C H A NG E
R E S İS T A N C E T O
C H A NG E
V alu e A d d ed
R is kin es s
B enc hm ar king
Im pr ovem ent in G u id anc e & C on tr ol
Im pr ovem ent in H um an R es ourc e Q u ality
Im pr ovem ent in P r od uc t & S er vic es
C r eativity
R es is tanc e
DATA TYPE
All variables collected are transformed into
dichotomous data, like change expected,
not expected; competition exists, do not
exist etc.
CONCLUSION
If business inputs* are forcing an
organization to change, the expectation of
employee development change is 90%.
In addition if benchmarking is
applied as a change means then this
percentage increases to 95%.
*like customer demand, bargaining power of customers &
suppliers, information and production technology)
CONCLUSION
But if the organization is not forced by
business inputs even then there is a chance of
change expectation if improvement in
guidance & control * is applied (78% expect
change).
This increases to 92% with the presence
of
force
of
laws
&
regulations
*like improvement in reward system, communication between
departments, quality control, & internal control
CONCLUSION
When there is no force of laws &
regulations if benchmarking is applied,
the change expectation rate is 80%.
CONCLUSION
Which emphasizes the importance
of benchmarking in change process.
Even when there is no force to
change if the organization is applying
benchmarking (which is actually a
proactive change strategy) even this is
enough to trigger change expectation.
CONCLUSION
On the other hand if there is no
force of business inputs, there is no
improvement in guidance & control, and
no force of competition then 82 % of
employees don’t expect to have a chance
to improve themselves.
CONCLUSION
As can be seen from the above
example every path has an implication.
IMPLICATION
What makes this study different
from other applications is the nature of
the problem explored. Decision tree
analysis are widely used in classification
of customers for segmentation purpose
and other CRM applications.
IMPLICATION
However the main purpose in this
study is to identify important variables
in
change
expectation
through
classifying the respondents on the basis
of their perceptions about the change
criterion.
IMPLICATION
Therefore by identifying these
respondents on the basis of the factors
effecting their change expectations, and
describing the important variables is a
valuable information for developing
strategies and policies of organizational
change.
Download

PowerPoint Sunusu