Scribd
Upload
11 views3 pages

Understanding Variates in Statistics

A variate is a random variable that can take on different values with a specified probability. Estimation error is the difference between the actual and estimated parameter values. Specification error occurs when the wrong model is estimated, such as when independent variables are incorrectly assumed to influence a dependent variable. Reliability analyses found Cronbach's alpha values above 0.7 for scales measuring knowledge integration management, information and communication technology, transformational leadership, and transactional leadership, indicating reliable data. Inter-item correlation tables for the knowledge integration management and information and communication technology scales showed no multicollinearity between items and sufficient convergence, supporting the validity of the scales.

Uploaded by

Ramzan Kaleem
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
11 views3 pages

Understanding Variates in Statistics

A variate is a random variable that can take on different values with a specified probability. Estimation error is the difference between the actual and estimated parameter values. Specification error occurs when the wrong model is estimated, such as when independent variables are incorrectly assumed to influence a dependent variable. Reliability analyses found Cronbach's alpha values above 0.7 for scales measuring knowledge integration management, information and communication technology, transformational leadership, and transactional leadership, indicating reliable data. Inter-item correlation tables for the knowledge integration management and information and communication technology scales showed no multicollinearity between items and sufficient convergence, supporting the validity of the scales.

Uploaded by

Ramzan Kaleem
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Variate

a Variate is a quantity which may take any of the values of a specified set with a specified
relative frequency or probability. The Variate is therefore often known as a random variable.

Estimation error is the difference between actual parameter and estimated parameter.

Specification error is actually a nice way of saying that the “wrong model” has been
estimated. More precisely, we assume that in the population there is a specific way in which
some set of independent variables, Xl to Xk, influence a dependent variable, Y. Specification
error can result in two ways.

AMERICAN PSYCHOLOGICAL ASSOCIATION (APA VERSION 7)

DATA ANALYSIS

Reliability Analysis

Table 1
Reliability of the Variables
Variable Cronbach's Cronbach's Alpha Based on No of
Alpha Standardized Items Items
Knowledge Integration Management .809 .806 12
Information and Communication
.862 .864 10
Technology
Transformational Leadership .895 .896 12
Transactional Leadership .814 .814 6

From Table1, the Cronbach’s alpha for each of the variable is greater than .7 indicating that

data for KIM, ICT, TRNSL, TRANL is reliable and can be used for further data analysis.
Inter-Items Correlation
Table 2
Inter-Items Correlation for Knowledge Integration Management

KIM1 KIM KIM3 KIM KIM KIM6 KIM KIM KIM9 KIM1 KIM1 KIM12
2 4 5 7 8 0 1

KIM1 1

KIM2 .461 1

KIM3 .357 .597 1

KIM4 .258 .367 .488 1

KIM5 .199 .237 .202 .347 1

KIM6 .135 .180 .172 .232 .584 1

KIM7 .142 .084 .097 .141 .338 .472 1

KIM8 .122 .068 .145 .132 .279 .285 .554 1

KIM9 .141 .087 .141 .202 .279 .349 .553 .521 1

KIM1 .136 .054 .035 .112 .209 .223 .240 .412 .408 1
0

KIM1 .178 .126 .092 .110 .239 .194 .352 .297 .398 .407 1
1

KIM1 .200 .133 .138 .104 .239 .322 .260 .289 .316 .308 .536 1
2

Referring Table 2, none of the item has higher correlation, i.e., r > .8 with any other item of

the same construct. Thus, there is no multicollinearity or duality of the items measuring the

construct. Moreover, all of the items have sufficient minimum level of correlation, i.e., r > .3

with other items of the construct. Therefore, it can be concluded that there is convergent validity

of the scale / items measuring the same construct.


Table 3
Inter-Items Correlation for Information and Communication Technology

ICT ICT
1 ICT2 ICT4 ICT5 ICT6 7 ICT8 ICT9 ICT10 ICT3
ICT1 1.000
ICT2 .633 1.000
ICT4 .262 .346 1.000
ICT5 .264 .281 .549 1.000
ICT6 .274 .287 .446 .624 1.000
ICT7 .311 .355 .319 .498 .594 1.000
ICT8 .177 .273 .334 .432 .556 .599 1.000
ICT9 .174 .191 .316 .396 .477 .511 .650 1.000
ICT10 .153 .178 .301 .433 .363 .437 .524 .634 1.000
ICT3 .364 .417 .428 .348 .388 .302 .383 .370 .326 1.000

From Table 3 it is concluded that none of the item has higher correlation, i.e. r > .8 with other

items of the same construct. It is therefore, it is concluded that there is no multicollinearity or

overlapping among items measuring ICT. It is also concluded that each of the item is correlated

with other items of the construct leading to conclude that there is convergent validity of the

scale.

You might also like