PBR 1007 Complete
-
Upload
adnan-jahangir-jutt -
Category
Documents
-
view
237 -
download
0
Transcript of PBR 1007 Complete
-
8/12/2019 PBR 1007 Complete
1/201
PAKISTAN BUSINESS REVIEW JULY 2010
Pakis tans Balance of Payments Prospects and Policy Proposals Motif
PAKISTANS BALANCE OFPAYMENTS PROSPECTS AND
POLICY PROPOSALS
M. Ashraf Janjua
Institute of Business Management, Karachi
Analysis of current account balance of Pakistan
As we look at the accounts on a long term basis we find
that since FY03, the trade balance in goods is in continuous
deficit and the deficit is ever increasing. It reached a maximum of
US $ 14,970 million in the year FY08 mainly because of increase
in oil prices (see next page).
The world economic slowdown and reduction in oil
prices brought down the trade deficit but still it remained
unreasonably high as compared to the potential of the economy.
The decreased deficit of the services account is mainly
attributable to the lower payments on account of transportation
because of lower imports, and the other major cause being the
lower payments for other business services. If we look at the
income account, there is an ever increasing debt burden.
Although foreign investment both direct and portfolio isattractive investors want return on their investment to stay in
business. Under the present circumstances it is a pipe-dream to
produce a positive income account. The only component of a
surplus is current transfers from workers abroad to their families.
As a rule of thumb, when a country is in a crisis, creditors avoid
lending any money or alternatively they also charge for the risks
involved. Such a situation gives rise to a vicious circle where a
country has low productivity, low exportable surplus, low
reserves for payment, low investment from abroad, low capacity
to borrow on soft terms and so on. In order to get out of this
vicious circle a country has to make extraordinary efforts. With
ever increasing current account deficit, we are facing difficulties
223
Motif
-
8/12/2019 PBR 1007 Complete
2/201
Motif
PAKISTAN BUSINESS REVIEW JULY 2010
Pakistans Balance of Payments Prospects and Policy Proposals
224
-
8/12/2019 PBR 1007 Complete
3/201
PAKISTAN BUSINESS REVIEW JULY 2010
Pakis tans Balance of Payments Prospects and Policy Proposals Motif
in controlling the budgetary deficit for various reasons, and there
is a lack of good governance. Though government has taken
several steps on its own and also as a part of IMF conditionality
more steps are needed to control expenditure and enhance
government revenues.
The main cause underlying our balance of payments
difficulties is our inability to increase value addition to the raw
materials we export. The main underlying reasons are the high
cost of energy for the production of tradables, poor infrastructure,
and low capital investment in the modernization of the machinery
and equipment to generate internationally competitive products.
Heavy investment, strategic planning and above all the
determination of the government and the nation are required for
gaining a competitive edge in the global markets in order to achievea trade surplus. It is worth mentioning that due to certain factors,
like low production, low quality, income inelastic demand for our
products and weak image of the Pakistani traders in the global
market, Pakistan has not been able to utilize its quota in the United
States and the European Union (average quota realization has
been around 70 %)in the less liberal scenario of the past1. We
need to explore the area of services where we can perform better
with relatively little investment and easily acquire competitive
advantage over our rivals.
1WTO Regime and Its Impact on Pakistan, Syndicate No. 1o,
Civil Services Academy, Lahore, 31stCommon Training Programme
225
-
8/12/2019 PBR 1007 Complete
4/201
Motif
PAKISTAN BUSINESS REVIEW JULY 2010
Pakistans Balance of Payments Prospects and Policy Proposals
Services are the fastest growing sector in the global
economy, constituting more than 60% of GDP of many countries.
Services are the largest and most dynamic component of both
developed and developing country economies. Services
currently account for over 60 percent of global production and
employment. Services, such as telecommunications, banking,
insurance and transportation are strategically important for
enhancing overall economic efficiency, performance and growth.
With services liberalization we may access quality service
providers and as a labor-abundant country we can develop
capabilities to capitalize our human resource with massive training
and development programs and capacity building initiatives. This
would generate savings, faster innovations enhance
transparency and predictability with technology transfer and
optimum utilization of the work force.
The following factors seem to be affecting our exports:
i. Law and order and war on terror affecting, among
other things, the inflow of foreign direct investment
(FDI)
ii. Power shortage also affecting investment flows, both
foreign and domestic.
iii. Erosion of competitiveness because of the increase in
unit prices of imports used as inputs for exportables.
iv. On the demand side constraints are.
a. Recession in the world economy althoughnow there are some indications of an upturn.
b. Dependence on raw material exports with low
value added.
c. A number of other countries have competitive
edge in case of a number of commodities and
services.
d . We have not made any visible progress in
diversification of our exports mix.
226
-
8/12/2019 PBR 1007 Complete
5/201
PAKISTAN BUSINESS REVIEW JULY 2010
Pakis tans Balance of Payments Prospects and Policy Proposals Motif
Around 30-35% of our imports are composed of crude/
furnace oil. We have to spend substantial amount on import of
edible oil, chemicals and chemical products for manufacturing
and agriculture, import of fertilizers, sugar and even wheat.
Workers remittances are a big part of net transfers and
it is this component that is largely supporting the balance of
payments along with other private and official transfers. As our
economy has the tendency of an increasing current account deficit,
we are facing difficulties in financing this deficit. War on terror
and expenditure on law and order is increasing our budget deficit
and building pressures on our external resources because we are
not generating enough public revenue. As a result the cost of
borrowing is increasing; FDI inflows are drying up and exerting
further pressures on cost of external financing. Our reserves havemainly been built on borrowing from the IMF.
Policy Proposals
Balance of payments problems may be resolved by
taking the following steps:
i. Efforts should be made to restore law and order and
conclude the countrys war on terrorism. Both these
factors are expected to pave the way for increase in
production and exports. Also, these improvements
should have salutary effects on FDI and tourism inflows.
ii. Increase in FDI should lead to inflow of advancedtechnology, expansion in services and growth in
production
iii. Inflow of FDI should also lead to capacity building in
human resource sectors with healthy effects on
production. Even if skilled people move abroad, that
should lead to expansion in home remittances.
iv. In addition to improvement in law and order, visible
progress should be ma de in good governance:
part icularly transparent and timely decision making,
monitoring of efficient implementation and setting up
effective accountability mechanisms.
227
-
8/12/2019 PBR 1007 Complete
6/201
Motif
PAKISTAN BUSINESS REVIEW JULY 2010
Pakistans Balance of Payments Prospects and Policy Proposals
v. Control of law and order situation will give a boost to
our stock exchange and consequently lead to an
increase in foreign portfolio investment.
vi. The quality of social and physical infrastructure should
be improved
vii. Alternative energy resources should be developed,
leading to enhanced production including exportable
surplus with increase in competitiveness
viii. There should be improvement in macroeconomic
stability, particularly the containment of inflation to a
modest level and keeping the Pakistan Rupee
competitive in the international market (in terms of
REER)
ix. Serious efforts are needed to diversify our exports withemphasis on the services sector, dairy products, fruitand vegetables and labor intensive segments of smallscale industry.
x. In addition to diversification of exports there is a need
to revisit the direction of trade and exploiting untapped
markets (including those in Africa and Latin America)
The long term sustainable level of the balance of
payments deficit depends on two fundamental variables: (1) the
ratio of foreign savings to investment and (2) growth in foreign
exchange earnings from exports of goods and services, workers
remittances and other private transfers. Depending on these two
variables, sustainable annual current account balance of
payments deficit could fall anywhere in the range of 2-3% ofGDP. Hence, there is need for developing guidelines and a
framework which will keep the current account balance of
payments deficit at sustainable levels taking into account the
gap between savings and investment and the growth in foreign
exchange earnings. Consideration may be given to the following
guidelines:
Establish a ceiling for the share of foreign savings in
total investment to ensure that large balance of
payments deficits do not finance consumption and that
the country doesnt become over-reliant on external
financing. Various empirical studies suggest that foreign
228
-
8/12/2019 PBR 1007 Complete
7/201
-
8/12/2019 PBR 1007 Complete
8/201
Motif
PAKISTAN BUSINESS REVIEW JULY 2010
Pakistans Balance of Payments Prospects and Policy Proposals
Investment (BOI) needs to gear up its efforts to attract
direct foreign investment in export related industries and
in the energy sectors.
It seems that competitive pressures for Pakistans textile
and clothing exports (64% of exports) arising from the
phasing out of the Multifibre Arrangement (MFA) were
not anticipated fully and new investments on the scale
required to move up the value chain have not been
forthcoming. The textile industry, thus, needs to adjust
to the realities of the world market and restore its
competitiveness through enhancing productivity. The
structural problems characterizing Pakistans textile and
clothing sector cannot be solved without major
investments in both plant and equipment and human
skills, investments in there areas are not taking place at
the required scale.
The government has already met the textile industry at
least half way largely through credit subsidies. The textile
industry needs to adjust to the realities of the world
market and must restore its competitiveness mainly
through enhancing productivity. Conceptually, any
subsidy in support of textile exports where international
prices are fall ing is not a good option. The fact that
Pakistani textile exports are dominated by cloth and yarn
while etc in the world clothing imports that are
expanding faster is a handicap which cannot be easilyaddressed by further cash subsidies.
Pakistans export sector, in general, is faced with
structural constraints e.g. extremely narrow export base,
low unit value exportables, lack of competitiveness, lax
quality control etc. Hence, any future export growth
strategy will have to be premised on structural solutions.
Any cosmetic measures to boost exports might yield
some marginal increases but would in no way address
the root causes of the problems afflicting the export
sector.
230
-
8/12/2019 PBR 1007 Complete
9/201
PAKISTAN BUSINESS REVIEW JULY 2010
Pakis tans Balance of Payments Prospects and Policy Proposals Motif
The government needs to carefully examine the
adequacy of the existing incentive regime for Pakistans
exports.
The government also needs to review the overall
production st ructure of the country to de termine
whether the existing tariff regime encourages
production for domestic consumption or for exports.
There is also a dire need to re-orientate our exports
strategy from the goods sector to services exports.
Services are a fast growing sector of the Pakistan
economy and their export potential needs to be tapped
through seeking enhanced market access.
The government should also assess the efficacy of
subsidies as a tool of export promotion and in this
regard an analysis ought to be made of the impact of
R&D support being provided to textile garments. The
evidence so far leads to the conclusion that financial
support has not helped expand value of textile exports.
For the long run an ambitious program of increasing
Pakistans market share in world trade which at present
is a paltry 0.15% is needed, but this cannot be achieved
without massive efforts to diversify exports and make
export development a central plank of our policy. Despitethe fact that Pakistan has liberalized its trade
substantially since the late 1980s (trade GDP ratio
changed to 0.5 during liberalized period from 0.1 during
the pre-liberalized period) it shows a much lower
response to trade liberalization in terms of exports
growth compared to its competitors and regional
counterparts (India; 0.4%; China 1.3%; Bangladesh
0.6%; Sri Lanka 0.3%; Malaysia 2.7%; Turkey 0.8% and
Iran 0.1%).
Policy attention needs to be focused on non-textile
manufactured exports, the new promising areas of IT
231
-
8/12/2019 PBR 1007 Complete
10/201
Motif
PAKISTAN BUSINESS REVIEW JULY 2010
Pakistans Balance of Payments Prospects and Policy Proposals
exports and agricultural and livestock products areas
where Pakistans world presence is minimum.
In order to broaden the export base, Government policy
must specially target foreign investment in
manufacturing, aimed both at improving technology and
productivity of promising export sectors.
A cabinet level committee to be chaired by the Prime
Minister could help improve much needed policy
coordination and implementation of export and industrial
policies as well as close monitoring of exports.
Import of goods management with competitive
alternatives, exports oriented imports, rails, roadsfacilities and other transport facilities, reducing wear
and tear leading to save imports on parts, domestic
savings and lower dependency on external resources
Imports of services management by developing our HR
capacity for earning of foreign exchange through
various modes of supply of services or otherwise
through migration of HR and earning transfers through
deploying our resources on more remunerative jobs.
We need to increase domestic savings and investment
and deploy our resources for financing a competitive
debt and equity mix so that we move to sustainability inthe medium term
Government policies need to be adjusted to ensure social
safety nets for the poor to protect the welfare of those
who are likely to be displaced in the transition period by
the process of trade liberalization and globalization more
generally.
Skill development and training schemes must be
instituted so that any displaced workers can be quickly
retained, relocated and reabsorbed in the labor force. It
is important to keep inflation in check in order not to
232
-
8/12/2019 PBR 1007 Complete
11/201
PAKISTAN BUSINESS REVIEW JULY 2010
Pakis tans Balance of Payments Prospects and Policy Proposals Motif
lose the gains from trade liberalization and other policy
reforms that have been undertaken.
Reduced tariffs, particularly on imported raw material,
imports, components and machinery can help boost
exports. However, other complementary policies are also
required. Meanwhile, with capacity constraints being
reached in the economy and inflation remaining high,
overestimating potential growth of the economy in the
short term runs the risk of letting demand grow at a rate
that cannot be sustained, which would make it difficult
to contain inflation.
Policy coordination on exports needs substantial
improvement. The responsibilities for export promotionare very dispersed among many agencies, including the
Ministry of Commerce, the Trade Development Authority
(a very good idea), the Ministry of Industry, the Textiles
Ministry, etc. A cabinet-level committee could help by
closely monitoring exports and speedily resolving policy
and implementation issues.
233
-
8/12/2019 PBR 1007 Complete
12/201
PAKISTAN BUSINESS REVIEW JULY 2010
ResearchProcess Capability Analysis for Non Normal Data
PROCESS CAPABILITY
ANALYSIS FOR NON
NORMAL DATA
Ejaz Ahmed
College of Computer Science andInformation Systems, Karachi
Suboohi Safdar,
Department of Statistics
University of Karachi, Karachi
234
Research
Abstract
Process capability analysis refers to the normal behavior of a
process when operating in a state of statistical control. Drives to
continuous improvement are usually associated with the processcapability measures. Typically we assume that the processes
follows normal probability distribution ensuring a high
percentage of the process measurements falling between 3of the process mean and the total spread amounts to about 6
variations. This article describes the estimation of pkp CanC d ,
commonly used process capability indices (PCI), in case of non-
normal data using the characteristics of Weibull distribution.
Earlier work of Lovelace and Swain (2009) has been extended for
this distributional assumption. Quantiles are estimated by
probability plotting technique and then control limits are obtained
to determine whether the process is in statistical control or not.Percentage points of the fitted distribution have been used to
compute under the assumption of Weibull distr ibution. We have
used Delta method (Stuart and Ord, 1987) to estimate parameters
and their standard errors. These estimated parameters are then
used to develop new PCIs. Average PCI values are given along
with the standard errors.
Keywords: Process capability, Estimation, Non normal Data,
Weibull Distribution
JEL Classification: C1160
-
8/12/2019 PBR 1007 Complete
13/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Process Capability Analysis for Non Normal Data
1. Introduction
Process capability refers to the inherent ability of a
process to produce homogeneous parts for a sustained time
period under given conditions. Kane (1986) described six areas
of applications of capability indices that include presentation of
non conforming products, measuring continuous improvement,
communication, prioritization, identifying direction for
improvement and auditing a quality system. Deleryd, et. al. (1999)
identified six critical factors for successful implementation of
process capabi lity studies. Tsim (1997) identi fied four key
objectives of PCI including, among others the ability to compare
different processes (unit less measure) and to identify the
closeness to target (Taguchi Loss Function concept).
Process capability measures the variability of a process
relative to its specification limit based on three assumptions,
namely, (i) the process is itself in control, (ii) target value and
specifications of a quality characteristics are specified, and (iii)
the process measures quality characteristics that follow a normal
distribution.
2. Process Capability Indices- Review
The most commonly used PCI, named pC Index,
measures the potential process performance (process
consistency) which only reflects the consistency of the productquality characteristic. pC measures process spread related to
specification limits and hence the location of process mean is
not considered. Assuming a quality characteristic to follow
( )2,N with upper and lower specification limits USL andLSL, this measure is defined as ( ) .6/ LSLUSLCp = The
expected proportion of nonconforming products assuming
( )2/)USLLSL+= can be obtained as
235
-
8/12/2019 PBR 1007 Complete
14/201
PAKISTAN BUSINESS REVIEW JULY 2010
ResearchProcess Capability Analysis for Non Normal Data
Where ( ). denotes standard normal cumulative distributionfunction and 2/)( LSLUSLd
= . Constable and Hobbs
(1992) defined capable as percentage of output within
specification while Montgomery (2001) recommended minimum
Cp equals to 1.33, for an existing process, and 1.50 for a new
process. Small values of Cpare bad sign but large values do not
guarantee of acceptability in the absence of information about
the values of the process mean.
As is obvious pC depends on the true standard
deviation (SD) of the process which is usually unknown. It
therefore forces to use an estimator of the SD which then results
in estimated PCI, based upon sample observations. The estimatedPCI given as is evaluated using
suitable unbiased estimator of , such as (Montgomery,
2001). An important ratio is quite frequently used as it
yields the statistics. This relationship
is useful in constructing confidence intervals or testing
hypotheses. Kane (1986) introduced another PCI known as
p kC which depends on both mean and standard deviation of the
236
( ) 6/ LSLUSLCp = 2dR
pp CC
( ) pnp CnC
=
211
( ) ( ) ( )
( )
[ ]
d
USLLSLLS L
LSLX
USLZpLSLZ
USLXpLSLXpE
=
=
+=
+ +
-
8/12/2019 PBR 1007 Complete
15/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Process Capability Analysis for Non Normal Data
process to deal with violation of centering assumptions and to
measure the actual capability performance. Mathematically this index
is described as .
Obviously, . Under
the assumption of normality exact confidence interval
for involves the joint distribution of two random variables
following non-central t-distribution. Nagata and Nagahata (1992)
showed that approximate confidence interval
for is given by,
These measures are unit less and permit comparison amongst
hundreds of process emanating from a whole range of productionprocesses and indust ries . However the methodology and
inferences about the process capability indices do not remain
too straight forward in the absence of normality assumption.
Next sections of this article will discuss the PCIs when the
underlying distribution is skewed.
3. Case under Non-Normality
Numerous authors have discussed the construction and
interpretation of PCIs under non-normal process behavior. Chen
at al. (1988) proposed PCIs with distribution free tolerance
intervals to estimate while Clement (1989) and McCormacket. al. (2000) proposed empirical non-normal percentiles to
evaluate both Cp
and Cpk
. Lovelace and Swain (2009) discussed
the construction of Cp
and Cpk
assuming process behavior
following a Log-Normal distribution. They used 99.875th and
0.135thquantiles to estimate both PCIs. Their proposed capability
indices are given below,
237
( ) ( ){ } 3/,3/ LSLUSLMinCpk =( )( ){ } ppk CdUSLLSLdC /2/+=
pkC
( ) %1001 pkC
)1(2
9
12/
2
+n
C
npkpk
ZC
-
8/12/2019 PBR 1007 Complete
16/201
PAKISTAN BUSINESS REVIEW JULY 2010
ResearchProcess Capability Analysis for Non Normal Data
Chen and Chen (2004) compared these two process
capability measures using four approximating methods, including
bootstrap estimators. Since skewed distributions are not too
common in production or service industries comparatively smaller
proportion of literature is devoted to address PCIs in non-normal
behaviors. Lovelace and Swain (2009) discussed both capability
indices assuming process data following a Log-Normal
distribution. Pal (2005) assumed process distribution to follow
Generalized Lambda distribution and evaluated PCIs. We intend
to discuss the construction of PCIs when the process distribution
follows a Weibull distribution. Interested readers are recommended
to refer to Munechika (1992), Pyzdek (1992), Kotz and Johnson
(1993), Somerville and Montgomery (1996), and Pal (2005) for
further details.
4. Process Capability Index- Weibull Distribution Case
Weibull distribution was originally proposed to describe
data from life testing commonly used in reliability, fatigue and
survivor analysis. Weibull distribution is a very important
distribution and has been widely used in studies related to, for
example, earthquakes, flood, breaking strengths, and reliability
under censored or truncated situations. However parameter
estimation is not easy especially in case of three parameter Weibull
distribution. Quantiles are recommended to be used while
estimating parameters. In this article we used delta method (Stuart
and Ord, 1987) to find the standard error of the estimates. Later
these estimated standard errors are used to construct our proposed
process capability index.
238
{ }
00135.0
99865.0
00135.099865.0
.
XMedian
LSLMedianC
MedianX
MedianUSLC
where
CCMinCXX
LSLUSLC
pl
pu
plpupk
p
=
=
=
=
-
8/12/2019 PBR 1007 Complete
17/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Process Capability Analysis for Non Normal Data
239
The three-parameter Weibull distribution probability
density function is given by
parameter Weibul l di stribution by assuming the location
parameter equal to zero. Many distributions are special cases
of Weibull distribution, for example exponential distribution is a
transformed form of Weibull distribution with shape parameter
.
We consider the two-parameter Weibull distribution with
probability function given as ,where x>0. A linear regression model was developed using the
cumulative distribution function, as described below.
The last expression (equation 2) is equivalent to
simple linear regression model, , where
Least square estimators of both parameters were determined andstandard errors were obtained using the Delta method (Stuart
and Ord, 1987) as given below:
( )
=
xxxf exp
1
where ,>x 0,0,0 >>> . We may derive the two-
1=
( ) ( )[ ]( ) ( ) ( )( ){ }[ ]
( ) ( ) ( )( ){ }[ ]
xFx
xFx
xxF
+=
=
=
1lnlnlnln
1lnlnlnln
exp1
ezy ++= 10 ( ) ( ) ( )( ){ }[ ]xFxy ==== 1lnlnzand,1,ln,ln 10
( ) [ ] ( ) [ ]( )
( ) ( ) ( ) ( )( )
)4(11111
)3()exp()exp()exp()(
2
2
4
1
1
4
1
1
2
2
1
1
2
1
2
2
2
2
00
2
00
2
0
=
=
=
=
==
=
yyVVVV
yyn
yVVV
i
i
i
......2
( ) ( )[ ]
xxF = exp1
-
8/12/2019 PBR 1007 Complete
18/201
-
8/12/2019 PBR 1007 Complete
19/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Process Capability Analysis for Non Normal Data
241
Table 1: Table showing PCIs based on Weibull Parameters
Alpha Beta se(Alpha) se(Beta) Cp(Alpha) Cp(Beta)
2 3 Mean 1.678 81.970 0.140 0.001 1.225 279.682
SE 0.141 214.749 0.024 0.001 0.219 125.525
3 3 Mean 2.417 144.614 0.327 0.000 0.761 69807.179
SE 0.239 248.027 0.458 0.000 0.208 254108.036
4 3 Mean 3.287 250.767 0.749 0.000 0.551 104157.087
SE 0.356 753.433 0.990 0.000 0.374 394228.414
5 3 Mean 3.375 104.995 0.875 0.000 0.451 14053.322
SE 0.315 305.475 1.021 0.001 0.251 53594.197
Control charts for with 6 variations can be easily plotted
constructing ( )[ ] 3 seUCL += and
( )[ ] 3 seLCL = . Table 1 reveals that the estimated
values for are highly deviated from the original parametric
values. The proposed method is recommended when the
process behavior follows a Weibull distribution and the
characteristic of interest is .
-
8/12/2019 PBR 1007 Complete
20/201
PAKISTAN BUSINESS REVIEW JULY 2010
ResearchProcess Capability Analysis for Non Normal Data
242
6. References
1. Chan, Lk., Cheng, S. W., and Spiring, F. A. (1988). A New
Measure of Process Quality: pmC . Journal of Quality
Technology, 20, 162-175.
2. Chen, J . P. and Chen, K. S. (2004). Comparing the
capability of two processes using Cpm. Journal of Quality
Technology, 36, 329-335.
3. Clements JA (1989) Process Capability Calculations for
non normal distributions. Quality Progress, 95-100
4. Constable, G. K and Hobbs, J. R, (1992), Small Samplesand Non Normal Capability, Trans, ASQCQuality
Congress 1-7.
5. Deleryd, M., Deltin, J. and Klefsjo, B. (1999). Critical
factors for successful implementation of process
capability studies. Quality Management Journal, 6, 40-59.
6. Kane, V. E. (1986), Process Capability Indices,Journal
of Quality Technology, 18, 41 -52.
7. Kotz, S. and Johnson, N. L. (1993),Process Capabil ity
Indices, London: Chapman & Hall.
8. Kotz, S. and Johnson, N. L. (2002). Process Capability
IndicesA Review, 19922000. Journal of Quality
technology, 34, 2-19.
9. Lovelace, C. R. and Swain, J. J. (2009). Process capability
analysis methodology for zero bound, non-normal
process data. Quality Engineering, 21, 190-202.
10. McCormark, D. W. Jr., Harris, I. R., Hurwitz, A. M. and
Spagon, P. D. (2000). Capability indices for non normal
Data.Quality Engineering, 12, 489-495
-
8/12/2019 PBR 1007 Complete
21/201
-
8/12/2019 PBR 1007 Complete
22/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
CROSS VALIDATION OF
RYFF SCALES OFPSYCHOLOGICAL WELL-
BEING: TRANSLATION INTO
URDU LANGUAGE
Sadia Aziz Ansari
Department of Social ScienceCollege of Business Management, Karachi
244
Abstract
The present study aimed to investigate the structural validity ofan Urdu translation of 54- items Ryff scales of psychological
well-being including; (six sub-scales: self-acceptance, positive
relations, autonomy, environmental mastery, personal growth
and purpose in life). Analyses were based on data from 261 men
and women, with a mean age of 25.64 yrs between 1860 years.
The calculated internal item correlation coefficients of the
translated scales were Cronnbachs alpha= (0.853), and
standardized item alpha= (0.855) significantly higher than the
original Ryff scales. Besides confirming previously reported
findings correlation among six subscales range between (r= 0.57
to 0.70). The present findings demonstrate the adequacy of the
Urdu version of the Ryff scales as instrument for assessing
psychological well-being among males and females in Karachi.
Keywords: Psychological well-being; Ryff scales; Urdu
translation.
JEL Classification: Z0000
Research
-
8/12/2019 PBR 1007 Complete
23/201
-
8/12/2019 PBR 1007 Complete
24/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
However, clarifying the structure of well-being has
neither been easy nor straightforward as evidenced by the
multitude of conceptual approaches that exist. For instance,
Bradburn (1969) proposed that SWB is composed of two
components, whereas others state that SWB consists of either5five components, 6six components or 7seven components.
Despite the lack of agreement about the number of
dimensions that compose SWB, a number of investigators agree
that SWB contains a cognitive and an affective component.8The
affective component is best understood as a hedonic balance
constituting ones overall emotional tone determined by an
individuals level of positive and negative affect and the difference
between these emotional states (Bradburn, 1969). Further, hedonic
balance is the conceptual basis for the most well known instrument
that measures the affective dimension of well-being9
The basis of the cognitive component of SWB emerged
from studies examining adaptation to gain and recognise the
contentment, or life satisfaction, approach. The logic behind this
approach is that if one has a favorable evaluation in many life
domains, such an evaluation will lead to an overall positive outlook
on ones life and the experience of higher levels of SWB.
3affectivepart, refers to both the presence of positive affect (PA)
and the absence of negative affect (NA)4cognitive part is an information-based appraisal of ones life.5(Lawton, 1975)6(Neugarten et al., 1961; Ryff, 1989)7(Reker and Peacock, 1981)8
8(Andrews and Withey, 1976; Diener, 1984; Diener and Emmons,
1984; Liang, 19848, 1985; Lucas et al., 1996; Stock et al., 1986).9Affect Balance Scale (ABS; Bradburn, 1969).
246
-
8/12/2019 PBR 1007 Complete
25/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-being
Ryff s Scales of Psychologi cal Well-being
According to Ryff (1989), well-being is not composed
simply of positive affect, negative affect, and life satisfaction;
rather, well-being is best conceived as a multidimensional
construct made up of life attitudes. Based on tenets of humanistic
psycho logy, wi th such cons truc ts as pu rpose in li fe and
autonomy, Ryff centers attention on normative criteria for mental
health. The result is a means for assessing a persons level of
positive functioning and psychological well-being. Ryff (1989)
created the Scales of Psychological Well-Being (SPWB) based
on an integration of mental health, clinical, and life span
developmental theories. These dimensions are assumed to
measure all aspects of wellbeing and include self-acceptance10,
positive relations with others
11
, autonomy
12
, environmentalmastery13, purpose in life 14, and personal growth15Ryff, 1989).
Ryff and Keyes (1995) examined the structure of Ryffs six factor
model using Structural Equation Modeling. The model that best
fitted the data was one of six primary factors joined together by
a single higher order factor defined as well-being. Ryff (1989)
also performed factor analysis on the six subscales of the SPWB
and found highest factor correlation between self-acceptance
and environmental mastery (0.76), self acceptance and purpose
in life (0.72).
10Self-Acceptance = positive evaluations of oneself11Positive Relations with others = quality relations with others12Autonomy = sense of self- determination13Environmental Mastery = capacity to effectively manage ones
life and surrounding world14Purpose in Life = belief in a purposeful and meaningful life15Personal Growth = sense of continued growth and development
as a person
247
-
8/12/2019 PBR 1007 Complete
26/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
Subscales loaded on a general well-being factor with
the remaining four subscales loading on two other factors.
Specifically, personal growth, positive relations with others and
purpose in life loaded on a factor believed to represent new
dimensions of well-being with control (i.e., powerful others,
chance).
Central to this discussion Kozma et al. (1991) reported
that it is important to establish the construct validity of a measure
by examining the extent to which the presumed components
emerge in factor-analytic studies. It is also imperative that the
items making up a measure load on the appropriate factors.
Establishing the construct validity of a measure is one way to
establish the usefulness of a scale. Therefore, the primary aim of
this study is to establish the construct validity of the Ryffs SPWB54-item (Urdu version).
I t i s hypothesis that an Urdu version of SPWB would
validate the construct vali dity of Ryff scale.
Extensive research exists on the correlation of
demographic and other environmental factors with happiness.
These findings started with Cantrils (1965) study of 23,875 people
in 11 countries, the research of Bradburn (1969) and Campbell,
Converse, and Rodgers (1976) in the United States, and Ingleharts
(1990) analysis of Eurobarometer studies of 16 countries with
over 163,000 respondents. Veenhoven and colleagues (1994) later
reviewed 603 such studies from 69 countries. It is concluded that
demographic and environmental factors affect happiness at
varying levels. Thus, fur ther thi s study will provide in sight on
subjective psychological well -bein g in Pak istani context.
248
-
8/12/2019 PBR 1007 Complete
27/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-being
Methodology
Ryff Scale of Psychological Well-being (RSPWB)
TheRyff Scale of Psychological Well-Being (RSPWB )
54-item, is a theoretically-grounded instrument that focuses on
measuring six dimensions of psychological well-being: self-
acceptance, personal growth, purpose in life, positive relations
with others, environmental mastery, and autonomy (Ryff, 1989).
Each dimensional scale contains 9 items equally split between
positive and negative items. Items are scored on a 6-point scale
ranging from strongly agree to strongly disagree.
Ryffs scales have been found to correlate positively
with prior measures of well-being, such as the
16
Affect BalanceScale and the 17Life Satisfaction Index. However, it is negatively
correlated with measures of depression like 18Zungs Depression
Scale. Internal consistency coefficients (alpha) for 19Ryffs six
sub scales range from (0.82 to 0.90).
Translation of Ryff Scale into Urdu
Dr.Carol Ryff, consented the author to translate the scale
into Urdu. Translation from English to Urdu and cultural
adaptation of scale was performed in two steps. The scale was
first translated by a bilingual expert working as assistant
professor; Communication at Institute of Business Management.
In the second step scale was examined by a Native Languageexpert working as lecturer at University of Karachi to avoid
syntactic errors in translation.
16(Bradburn, 1969)17(Neugarten et al., 1961)18Zungs (1965)19(Schmutte and Ryff, 1997).
249
-
8/12/2019 PBR 1007 Complete
28/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
Sample:
Participants were university students, un/married, un/
employed individuals in the vicinity of Karachi city. Three hundred
participants completed questionnaires for the study. .Thirty nine
questionnaires were excluded for data analysis due to incomplete
responses. Two hundred sixty-one research participants made
up the final sample. The sample contained 111 males and 150
females, whose mean age was 25.43 yrs (SD = 3.76; range: 1848
years) and mean education of 14.22 years (SD = 1.37; range: 10
19 years).
Data Collection:
All respondents were asked to fill the RSPWB 54-item
Urdu Version questionnaire either in groups or individually.
Group administration of RSPWB 54-item was undertaken
at the Institute of Business Management (IoBM), Karachi
Foundation School (KFS) and Meezan Bank. 120 students from
the Freshman Introductory Psychology class at IoBM, 34 teachers
and management staff at KFS and 45 employees from Meezan
Bank were conveniently selected at random to complete the
questionnaire. 68 questionnaires were dispatched to postal
addresses in the vicinity of Karachi and the author received 24
completed questionnaires. The rest of the data was collected bydistributing the questionnaire in the locality of Karachi city.
Results
Demographic profiles of participant suggest that sample
was largely comprised of (73.3%) unmarried and (25.6%) married
individual from general population, with average age of 25.43yrs
between (18-60 years).A large group of respondent were between
17-25yrs of age (38.3%) while others were (19.1%). Educational
qualification of majority of research participant were grouped
into under graduation (Intermediate=28%,A-levels=13%) and (
Masters=31% ).On the other hand, the rest of the participant
250
-
8/12/2019 PBR 1007 Complete
29/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-being
were graduate (20%), professionals such as doctors, Engineers
(05%) and matric (03%). The gender profile of research participant
showed a bit larger female composition (57.6%) than male (42.4%).
The construct validity of RSPWB 54-item was tested
against RSPWBs Urdu Version. Estimated internal item
correlation coefficients of the translated scales were
(Cronnbachs alpha (=0.853), and standardized item alpha=
(=0.855) see table 1. Inter-correlation among six subscales range
between (r=0.57 to 0.70) correspondingly: autonomy and
environment mastery (r=0.54), autonomy and personal growth
(r=0.64), autonomy and positive relation (r=0.45), autonomy and
purpose in life (r=0.34), autonomy and self acceptance (r=0.53),
environmental mastery and personal growth (r=0.58),
environmental mastery and positive relations (r=0.54),environmental mastery and purpose in life(r=0.47), environmental
mastery and self acceptance (r=0.46), positive growth and positive
relations(r=0.55), positive growth and purpose in life (r=0.42),
positive growth and self acceptance (r=0.46), positive relations
and purpose in life (r=0.48), positive relations and self acceptance
(r=0.38), purpose in life with self acceptance (r=0.51) were
significant at 0.01 level (see table 3). Obtained value suggests
that probability of individual item responses on six subscales
were consistently same with the total responses (See Table 2).
Considerably low inter correlations depicted among
subscale of Autonomy and purpose in life scale20and positive
relations and self acceptance scale 21.This is contrary to Ryffssix factor model for subjective psychological well-being. It
remains debatable; Do the items intended to measure each
theoretical domain? 54-items are enough to measure subjective
psychological well-being? Is their an overlap of items pertaining
to more than one domain? This may recommend that Subjective
Psychological well-being in Pakistani context comprises of less
than six subscales.
20Autonomy and purpose in life scale (r=0.34)21Positive relations and self acceptance scale (r=0.38)
251
-
8/12/2019 PBR 1007 Complete
30/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
However considerably strong inter correlation among
subscales appeared between autonomy and personal growth
(r=0.64), and personal growth and environmental mastery (r=0.58).
These validate Ryffs theory of six factor psychological well-
being measure. Psychological well-being is greater, as majority of
respondent were unmarried undergraduates between 17-25yrs of
age.22 This also suggests dominate of factors that determine
psychological well-being in Pakistani context.
Ryffs theory of PWB equates autonomy with self-
determination, independence, internal locus of control,
individuation, and internal regulation of behavior. While other
authors assumed that autonomy is related to the western concept
of liberty and freedom. However present study suggests that the
attribute of autonomy also exist in non-western cultures. Such as
item 35 on RSPWB I have confi dence in my opini ons, even i f
they are contrar y to the general consensus reveals higher
consistency than other individual items on the subscale of
autonomy (r=0.46).
Likewise, it is reported previously that personal growth
is the ability to grow and expend as a person who is considered
as sense of individuals well-being rather than moral imperative.23
Even though inter correlation among all six subscales:
autonomy, environmental mastery, personal growth, purpose in
life, positive relations and self acceptance with total subjective
psychological well-being was strongly larger than the individual
items on six subscale. Inter correlation range between (r=0.70 to
r=0.79).22A Small but significant correlation between education and SWB
is indicated (Campbell et al., 1976; Cantril, 1965; Diener et al.,
1993). In a meta-analysis of the literature, Witter, Okun, Stock,
and Haring (1984) observed a median effect size of .13.This effect
size was similar to educations influence upon life satisfaction
(.15), morale (.15), and quality of life (.12). Education correlates
with well-being moreso for individuals with lower incomes
(Campbell, 1981; Diener et al., 1993), and in poor countries
(Veenhoven, 1994a);23 (Bellah et al. 1985, p.47)
252
-
8/12/2019 PBR 1007 Complete
31/201
-
8/12/2019 PBR 1007 Complete
32/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
References
Andrew, F.M and S.B Withey : 1976, Social Indicators of Well-being (Plenum Press , New York).
Bradburn, N.M 1969,The Structure of the Psychological Well-
being (Aldine Publisher Co, Chicago)
Diener, E. and R.A. Emmons: 1984, The independence of positive
and negative affect, Journal of personality and social psychology
47, pp. 1105-1117.
Diener , E, R.A., Emmons, R.J.Larsen, and S.Griffin:1985, The
Satisfaction with Life Scale, Journal of personality Assessment
49,pp.71-75.
Diener ,E,Suh, E.M, Lucas, R.E & Smith,H.L. (1999) Subjective
well-being: Three decades of progress. Psychological Bulletin,
125(2),276-302. Doi: 10.1037/0033-2909.125.2.276.
Liang J:1984, Dimensions of the life satisfaction Index A:A
structural Formulation, journal of Gerontology 39,pp. 613-622.
Lucas, R.E, E.Diener and E.Suh:1996, Discriminant validity of life
satisfaction, Journal of Personality and Socail Psychology 71,
pp.616-628
Neugarten,B.,R.Havighurst and S.Tobin:1961, The measurement
of life satisfaction, Journal of Gerontology 16,pp.134-143.
Reker, G.T. and E.J.Peacock:1981, The life Attitude Profile (LAP):
A multidiemensional Instrument for assessing attitude towards
life Canadian Journal of Behavioral Sciences 13, pp.264-273.
Ryff, C.D. 1989, Happiness is everything or is it? Explorations
on the meaning of psychological well-being, Journal of Personality
and Socail Psychology 57, pp.1069-1081.
Ryff, C.D. 1989, Beyond Ponce de Leon and life satisfaction:
New directions in quest of successful aging International Journal
of Behavioral Development, 12, pp.35-55.
254
-
8/12/2019 PBR 1007 Complete
33/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-being
Schmutte, P.S. and C.D. Ryff: 1997, Personality and well-being:
Reexamining methods and meanings Journal of Personality and
Socail Psychology 73, pp.549-559.
Stock,W.A.,M.A.Okun and M.Benin: 1986, Structure of
Subjective well-being among the elderly Psychology and
Aging 1, pp.91-102.
Correlations among six factor Model of Ryffs Psychological
Well-being Scale (54-item)
Table 1: Item analysis Statistics SPSS out put
Cronbach's
Alpha
Cronbach's
Alpha Based onStandardized Items
N of
Items
.853 .855 6
Table 2: Descriptive Statistics for Ryffs
Psychological well-being Six Subscales
Scales
Mean
Std.
Deviation N
Autonomy
37.72 6.022 261
*E.Master
y
37.13 6.334 26
1
**P.Grow
th
32.19 5.502 26
1
***P.Rela
tion
35.49 6.241 26
1
****P.I.L
ife
32.07 6.514 26
1
*****S.Accept
35.54 5.508 261
Note: (Ryffs Psychological Well-being Scale 54-item 6 factorModel (Autonomy,*E. Mastery =Environmental Mastery,
**P.Growth= Personal Growth, ***P.Relation=Positive Relation,
****P.I.Life= Purpose in Life, *****S.Accept= Self Acceptance)
(N=261, Age= 25.64yrs )
255
-
8/12/2019 PBR 1007 Complete
34/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
Table 3: Inter-Item Correlation among Ryffs Psychological well-being
Subscales
Scales Autonomy E.Mastery P.Growth P.Relation P.I.Life S.Accept TPWB
Autonomy 1.000 0.545 0.648 0.454 0.349 0.534 0.73
E.Mastery 0.545 1.000 0.584 0.549 0.478 0.469 0.77
P.Growth 0.648 0.584 1.000 0.557 0.429 0.460 0.79
P.Relation 0.454 0.549 0.557 1.000 0.485 0.382 0.73
P.I.Life 0.349 0.478 0.429 0.485 1.000 0.512 0.70
S.Accept 0.534 0.469 0.460 0.382 0.512 1.000 0.71
TPWB 0.73 0.77 0.79 0.73 0.70 0.71 1.00
Note: TPWB= Total Psychological well-being (all correlations are significant at 0.01 level)
Table 4: Inter-Item Covariance among Ryffs Psychological well-being
Subscales
Scales Autonomy E.Mastery P.Growth P.Relation P.I.Life S.Accept TPWB
Autonomy 36.265 20.776 21.466 17.063 13.684 17.723 123.0
E.Mastery 20.776 40.121 20.352 21.690 19.726 16.360 138.0
P.Growth 21.466 20.352 30.271 19.129 15.368 13.931 120.0
P.Relation 17.063 21.690 19.129 38.951 19.733 13.126 127.0
P.I.Life 13.684 19.726 15.368 19.733 42.430 18.353 129.0
S.Accept 17.723 16.360 13.931 13.126 18.353 30.342 110.0
TPWB 123.0 138.0 120.0 127.0 129.0 110.0 789.23
Table 5: Summary Item Statistics Ryffs Psychological well-being Scales SPSS out put
Mean Minimum Maximum Range
Maximum /
Minimum Variance
N of
Items
Item Means 35.022 32.065 37.716 5.651 1.176 5.789 6
Item Variances 36.397 30.271 42.430 12.160 1.402 26.202 6
Inter-Item
Covariances
17.899 13.126 21.690 8.563 1.652 7.973 6
Inter-ItemCorrelations
0.496 0.349 0.648 0.299 1.857 0.006 6
256
-
8/12/2019 PBR 1007 Complete
35/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-being
Table 6: Item-Total Statistic Ryffs Psychological well-being Scales SPSS out put
Scales
Scale Mean if
Item Deleted
Scale
Variance if
Item Deleted
Corrected
Item-Total
Correlation
Squared
Multiple
Correlation
Cronbach's
Alpha if Item
Deleted
Aoutunmy 172.42 537.652 .650 .515 .827
E.Mastery 173.00 517.412 .686 .478 .819
P.Growth 177.94 544.577 .703 .542 .818
P.Relation 174.64 534.907 .629 .427 .831
P.I.Life 178.07 539.180 .574 .388 .842
S.Accept 174.59 566.011 .607 .415 .835
((all correlation are significant at 0.01 level)
Appendix B: Ryffs Psychological Well-being Scale Urdu Translation
257
-
8/12/2019 PBR 1007 Complete
36/201
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-beingResearch
258
-
8/12/2019 PBR 1007 Complete
37/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Cross Validation of Ryff Scales of Psychological Well-being
259
-
8/12/2019 PBR 1007 Complete
38/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
260
DIFFERENCES IN MOBILE
SERVICE PERCEPTIONS:
COMPARISON OF STUDENTS
AND STAFF AT A BUSINESS
UNIVERSITY
Laiq Muhammad KhanCollege of Computer Science and Information Systems,
Karachi
Abstract:
Mobile services are widely used all over the world and with theincrease of mobile service users the competition between
different mobile service providers is also increasing in every
country. Mobile service users are very conscious about networkquality, perceived value, billing service, satisfaction etc.
Perception of users about different services and packages offeredby different mobile service providers varies age wise, gender
wise, country wise etc.
This study aims to investigate the difference in mobile serviceperception and its impact on perceived value, satisfaction, loyalty
between two significant groups of mobile service users, thestudent and staff of a big business university at Karachi.
A group of statistical technique comprising analysis of variance,regression and inferential statistics are used for testing
hypothesis about the attributes and for developing model of
loyalty.
The results identify the mobile service quality attributes that are
important for two groups of users. This study also findssignificant difference between the two groups of users in terms
of effect of perceived economic and emotional value, satisfaction,
network quality and loyalty.
Key Words: Customer loyalty, customer satisfaction, perceived
value, network qualityMobile service .
JEL Classification: M3310
-
8/12/2019 PBR 1007 Complete
39/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
261
I. INTRODUCTION
Mobile phones are widely used all over the world. The
number of mobile users around the world was 1.5 billion at the
middle of the year 2004 which is about 25% of the population
(Mobile Tracker: Cell phone Demand 2005). By the end of the
year 2005, it reached two billion (CNN 2005). In the year 2009,
40% of the world population were mobile phone users (Gartner;
2009)
According to Pakistan Telecommunication Authority
(PTA) Industry Analysis Report 2008 Cellular mobile services in
Pakistan commenced in the 90s with two mobile service provider
Paktel and Pakcom ( Instaphone ). There has been strong growth
in the cellular market. By the end of 2007; five cellular operators
were in the market in Pakistan.
The growth rate of number of subscribers was 80% in
the year 2007. The total subscribers were 76.9 million by Dec.
2007, it was 34.5 millions in the year 2006, and 12.7 millions in
2005.In 2008-09, the cellular mobile companies added over 6.3
million subscribers, while the previous year, the addition was
about 25 million. During the period 2007-08 the number of
subscribers were 88 millions, 94.3 millions during the period 2008-
09, about 97.6 millions by the end of year 2009 and 95.4 millions
by the end of Jan. 2010 (PTA report on March 11, 2010). In USA,
Mexico, Hong Kong, Taiwan, China studies have been made
about the difference in perception of mobile service between
users of different age group. We have not found any such study
about the users of mobile services in Pakistan.
The purpose of this study is to investigate the difference
in mobile service quality perceptions and its impact on perceived
value, satisfaction, and loyalty between two important mobile
service users groups i.e. students and staff at a leading Business
university at Karachi in 2010.
As the usage pattern of these two groups is distinct,
the life style of these two groups as well as the technology
diffusion may cause significant differences in their satisfaction
and loyalty decisions.
-
8/12/2019 PBR 1007 Complete
40/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
262
II. LITERATURE REVIEW
Kumar, A and Lim,H(2006)under took this study to
investigate the effects of age on mobile service quality perceptions
and its impact on perceived value, satisfaction and loyalty
decisions with respect to two different consumers groups 12 years
to 26 years termed as Generation y and 42 60years termed as
Baby boomers.
They have collected data for the age group 18-24 years
old from a sample of 159 out of the total population of the students
at Southern University USA, using the method of convenience
sampling. The data for the persons of the age group 42-60 years
has been selected on the basis of a sample size of 139 through a
web-based survey. In both the groups, the respondents are onlymobile services users.
In this study analysis has been performed in three steps.
1. Separate baseline models have been established
for both the groups Gen Y and baby boomers
by using the data sets for Gen Y and baby
boomers respective ly. The signif icant paths
were reconfirmed by using the multi-group SEM
model. Chi Square statistic has been used to
test the goodness of fit.
2. I n o rd er t o a sse ss t he m et ri c e qui va le nc yof the constructs in the two groups, a
measurement model has been fitted.
Goodness of fit test has been performed.
3. At the fi nal stage, to tes t the equa li ty of the
structural parameters additional constraints
were added and test1 of invariance has been
performed by using Chi- Square stat ist ic.
1They have used test of invariance to discover that the structural
parameters are equal or not across the groups of the babyboomers and Generation-y 1
-
8/12/2019 PBR 1007 Complete
41/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
263
They have tested several hypotheses.
About the effect of network qual ity on perceived
economic and perceived emotional value. They have
concluded that there is significant positive effect of
network quality on perceived economic and emotional
value for both groups.
About the effect of billing system on perceived
economic value and perceived emotional value for both
age groups. They have concluded that there is a
significant difference between billing system and
perceived economic and emotional value for the two
groups.
About the effect of counter service quality on perceived
economic value and perceived emotional value for thetwo groups. They have concluded that customer service
quality has an insignificant relationship with perceived
economic value and emotional value for baby boomers
while customer service quality has a positive effect on
perceived emotional value alone for Gen Y- ers.
About the effect of emotional value on satisfaction for
the two age groups. They have concluded that there is
significant difference for both the groups with respect
to emotional value and satisfaction; perceived emotional
value has a greater effect on Gen Y-ers than the baby
boomers.
About the effect of perceived economic value on
satisfaction for the two age groups, it has been observedthat there is a positive significant effect of perceived
economic value on satisfaction for baby boomers but
not for Gen-Y and hence the effect of perceived
economic value on satisfaction is greater for baby
boomers.
Relationship between Satisfaction and Loyalty has been
studied and it was found that there is a positive relationship
between satisfaction and loyalty in both age groups.
Yang, Z and Peterson, R.T (2004)have undertaken a
study to investigate the moderating effects of switching over
-
8/12/2019 PBR 1007 Complete
42/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
264
costs on customer loyalty through both satisfaction and perceived
value measures.
They have collected data through a Web-based survey.
Arandom selection of 4000 subjects from an e-mail list
provided by an e-mail broker has been selected and respondents
were invited to participate in the survey .Responses from 257
participants were received by the authors , including 22 incomplete
or duplicated responses,therefore the actual sample size is 235.The
analysis has been performed in four steps
1. In the first step, exploratory factor analysis has been
performed to determine the underlying factor structure of the scale
items. 2. In the second phase, confirmatory factor analysis has
been performed.
Chi-Square statistics have been used and goodness of fit
test has been performed in order to test the fitting ofthe model.
3. The discriminant validity of the measures has been
examined by two different procedures.
a) The AVE2has been compared with the square of the
parameter estimate among the latent variables.
b) The discriminant validity of each construct has beendetermined by loading higher on the construct of
nterest than any other variable.
4. Simultaneous maximum-likelihood-estimation
procedures have been used in order to examine the hypothesized
relationships among perceived, customer satisfaction, and
customer loyalty. Goodness of fit test has been performed.
They have tested the hypothesis:
About the effect of customer loyalty on customer
perceived value. As the difference is significant, therefore
-
8/12/2019 PBR 1007 Complete
43/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
265
it is concluded that customer loyalty will be positively
influenced by customer perceived value
About the effect of customer loyalty on customer
satisfaction. As the difference is significant, therefore it is
concluded that customer loyalty will be positively
influenced by customer satisfaction.
About the customer satisfaction on customer perceived
value. As the difference is significant, therefore it is
concluded that customer satisfaction is positively
influenced by customer positive value.
About the level of switching cost on customer loyalty
through customer satisfaction, As the difference is
insignificant, it is concluded that the higher the level of
switching cost customer loyalty will not lead to greater
customer satisfaction.
About the level of switching cost on the customer loyalty
through perceived value. As the difference is insignificant,
hence it is concluded that the higher the level of switching
cost, the perceived value will not lead to greater customer
satisfaction
Anderson, R.E and Srinivasan, S.S (2003)undertook a
study to observe the impact of satisfaction on loyalty in electronic
commerce (e-commerce). In order to observe the relationship between
satisfaction and loyalty, they included the variables convenience
motivation and purchase size as consumer level factors where as
trust and perceived values were included as business level factors.
2The AVE (Average Variance Extracted) represents the amount of
variance captured by the constructs measures relative to
measurement error and the correlations among the latent variables.
-
8/12/2019 PBR 1007 Complete
44/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
266
Initially they selected a random sample of 5000 consumers
from a large list of e- retailing consumers maintained by an online
marketing research firm. They collected the data by sending an
invitation through an e-mail to each of the 5000 respondents and 1211
complete and usable responses were received. The representative-
ness of the data has been evaluated by comparing the collected sample
data with the data collected by a Greenfield Online study showing
similar demographic characteristics. To measure various constructs,
validated items used by other researchers have been adapted.
For the purpose of analysis, the sample has been split in to
two sets:
(a) An exploratory data set of 360 observations.
(b)The model estimation data set of 851 observations.
1. An exploratory factor analysis technique has been applied to the
exploratory data set and internal consistency estimates were obtained.
High internal consistency between various constructs was observed
on the basis of the estimates.
2. The model for loyalty has been obtained on the basis of the model
estimation data by applying regression analysis. The coefficient alphas,
means and standard deviations for various constructs on the basis of
model estimation data sets have been obtained.
The regression model which was run in this research paper is as follows:
LT = 0+ 1 SA + 2TR + 3PV + 4IN + 5CM + 6SA*TR +
7SA*PV +
8SA*PS +
9SA*IN +
10SA*CM +
LT: e- Loyalty
SA: e- Satisfaction
TR: Trust in the e-Business
PV: Perceived value
IN: Inertia
CM: Convenience orientation
PS: Purchase size
-
8/12/2019 PBR 1007 Complete
45/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
267
Their tested hypotheses are:
About the effect of the level of e-satisfaction on the level
of e-loyalty. As the difference is significant, therefore it is
concluded that the higher the e-satisfaction, the higher
the e-loyalty
That the impact of customer e-satisfaction on e-loyalty is
moderated by inertia. As the difference is significant, it is
concluded that e-satisfaction will have a higher impact on
e-loyalty at a lower values of inertia than at higher values
of inertia.
That the impact of customer e-satisfaction on e-loyalty is
moderated by convenience motivation. As the difference
is significant, it is concluded that convenience motivation
positively moderates the impact of customer e-satisfactionon e-loyalty.
That the impact of customer e-satisfaction on e-loyalty is
moderated by purchase size. As the difference is
significant, it is concluded that purchase size moderates
the impact of customer e-satisfaction on e-loyalty.
That the impact of customer e-satisfaction on e-loyalty is
moderated by trust. As the difference is significant, it is
concluded that trust moderates the impact of customer e-
satisfaction on e-loyalty.
That the impact of customer e-satisfaction on e-loyalty is
moderated by trust. As the difference is significant, it is
concluded that trust moderates the impact of customer e-
satisfaction on e-loyalty. That the impact of customer e-satisfaction on e-loyalty is
moderated by perceived value of a Web site. As the
difference is significant, it is concluded that perceived
value moderates the impact of customer e-satisfaction on
e-loyalty.
Definitions of Variables (Kumar, A and Lim,H, 2006)
Service quality
Perceived value (i.e. economic, emotional)
Satisfaction and
Loyalty intention
-
8/12/2019 PBR 1007 Complete
46/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
268
Overall service quality:
Quality is defined as the overall excellence or superiority
that consumers perceive from a product/service
(Zeithaml, 1998). Service quality in the use of mobile
services can be perceived through both technical (e.g.
pricing plans, network quality and data services) and
functional attributes (e.g. billing system and customer
service quality) of mobile services. Overall perceptions
of service quality are formed by a consumers evaluation
of multiple quality dimensions (Gronroos, 1984). In
general, researchers, agree that positive perceptions of
service quality enhance consumers perceived value and
the level of satisfaction. In other words, a consumers
initial appraisal of service quality can arouse positiveemotion, which results in behavioral responses (Bagozzi,
1992).
Perceived Value:
Previous studies examined perceived value in terms of
monetary tradeoffs only (McDougall and Levesque,
2000) what you get for what you pay. However,
consumers appear to assets perceived value not only by
monetary tradeoffs but also by other psychological
benefits (e.g. enjoyment and fun) (Sweeney and Soutar,
2001). Similarly, previous studies emphasized both
intrinsic and extrinsic motivations as predictors ofbehavioral intentions (e.g. Davis et al.,1989). While
extrinsic motivation is goal oriented, intrinsic motivation
pertains to the pleasure and inherent satisfaction driven
by service experience (Vankatesh et al., 2000). Therefore
this study measures the effects of both perceived
economic and emotional value on consumers
satisfaction.
Satisfaction and loyalty:
-
8/12/2019 PBR 1007 Complete
47/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
269
Satisfaction has been widely studied as a mediator
between perceived value and loyalty (Andreassen and Lindestad,
1998; Cronin et al., 2000). Customer satisfaction is an important
factor for a long term relationship between a firm and a customer
(Anderson and Srinivasan, 2003). Loyalty refers to a consumers
commitment to repurchase a preferred product or service
consistently in the future (Oliver, 1980). Researcher has shown
that the consumers positive affect toward a service provider is
likely to motivate the consumer to stay with the provider and
also recommend the service to other (Zeithaml et al., 1996).
Therefore, this study measures the direct effect of satisfaction
on consumers loyalty decisions.
III. Methodology
The purpose of our research is to investigate the
difference in mobile service quality perceptions and its impact
on perceived value, satisfaction, and loyalty between two
important mobile service users groups i.e. students and staff at a
major Business University in Karachi. Consequently the group
of students is appearing as the group of consumers of mobile
service without income and the group of staff is the group of
consumers with income.
As the usage pattern of these two groups is distinct, the life
style of these two groups as well as the technology diffusion
may cause significant differences in their satisfaction and loyalty
decisions.
This study proposes that differences in mobile service usage
between students and staff may be caused by different attributes.
The following hypotheses have been tested.
H1a: customer loyalty will be positively influenced by customer
emotional value
H1b: customer loyalty will be positively influenced by customer
economic value
-
8/12/2019 PBR 1007 Complete
48/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
270
H2a: customer satisfaction will be positively influenced by customer
economic value
H2b: Satisfaction will be positively influenced by customer
emotional value
H3: Data services will have a greater effect on perceived economic
value for students than for staff.
H4: Data services will have a greater effect on perceived emotional
value for students than for staff.
These hypotheses are based on the literature survey
formulated in Section II above.
Measures
The research instrument has been adapted from the researches of
Kumar, A and Lim, H(2006) , Yang, Z and Peterson, R. T(2004),
Anderson, R. E and Srinivasan, S.S (2003)
The likert scale rating 5 steps has been used to measure the
variables
1=strongly agreed, 2=agreed, 3= dont know, 4= disagreed, 5=
strongly disagreed
But for the negative response questions Q.13, 14,15and Q.25 scale
rating is1= strongly disagreed, 2=disagreed, 3=dont know, 4=agreed,
5=strongly agreed
The questionnaire comprising of the questions regarding mobile
service in use, age, gender, perceived quality of mobile service,
perceived value, the level of satisfaction, and loyalty.
The questionnaires for both the groups comprises of 28 questions.
However the questionnaire for staff includes two additional
informations about designation and education.
-
8/12/2019 PBR 1007 Complete
49/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
271
To measure the various concepts, validated items used by other
researchers have been adapted.
To capture the consumers perception about Billing service two
items have been adapted from the study of Kumar .A and Lim
H(2006). Q.8 and Q.9 of this study are similar to that of Kumar A
and Lim H Questionnaire and measures the consumers
perception about billing service.
The consumers perception about the Network Qualityhas been
measured by three items. Two items have been adopted from the
study of Kumar .A and Lim H (2006), Q.10 and Q.12 are common
and Q.11 (see appendix II) about the Voice Quality has been
introduced in this questionnaire.
The Consumers perception about the Customer Service Quality
has been measured by using three items. These items have been
adopted from the studies of Kumar A and Lim H (2006), Yang Z
and Peterson R.T (2004). Q.13 and Q.14 and Q.15 (see appendix
II) of this study are measuring the perception about Customer
Service Quality.
Data servicesin this study have been measured by using two
items. These items have been adopted from the studies of Kumar
A and Lim H (2006), Yang Z and Peterson R.T (2004). Q.16, Q.17
and Q. 18 (see appendix II) of this study are common and
measuring the perception of consumer about Data services.
Perceived value includes both monetary and non-monetary
benefits that consumer perceive in a service setting. In this study,
we consider perceived value to include economical and emotional
value (Kumar .A and Lim H, 2006), Perception about theEconomic
valuehas been measured by using three items. Two items have
been adopted from the studies of Kumar A and Lim H (2006),
Anderson R E and Srinivasan SS (2003) Q.19 and Q.20 (see
appendix II) of this study are common and a new question Q.21
(Number of SMS you send per day) has been introduced.
-
8/12/2019 PBR 1007 Complete
50/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
272
Consumers perception about the Emotional value has been
measured by using three items. All items have been adopted from
the studies of Kumar A and Lim H (2006), Anderson R E and
SrinivasanS S (2003). Q.24, Q25, Q.26 (see appendix II) of this
study measures the consumers perception about Emotional value.
Consumers perception about the Satisfactionhas been measured
by using two items. Both items have been adopted from the studies
of Kumar A and Lim H(2006), and Anderson R E and SrinivasanS
S (2003). Q.22 and Q.23 (see appendix II) of this study measures
the perception about the satisfaction.
Consumers perception about the Loyaltyhas been measured by
using two items. Both items have been adopted from the studies
of Kumar A and Lim H (2006), Anderson R E and SrinivasanSS(2003). Q.27 and Q.28 (see appendix II) of this study measures the
consumers perception about the Loyalty.
In this study some questions different from the questionnaires of
other researchers are being included due to the change of the
environment, habits and behavior of the consumers. These
questions are about the causes of selecting mobile service,
changing mobile service, using more than one mobile service at a
time, and the type of package.
The question about the rating of SMS service is not included in
the questionnaire of the other researchers. (How will you rate
SMS service of your selected mobile company?)
This research is different from the study of Archana Kumar and
Heejin Lim (2006), Rolph E. Anderson and Srini S. Srinivasan(
2003 ) ,Zhilin Yang and Robin T. Peterson ( 2004 ) on the following
basis:
The groups are the students and the staff of a large educational
institution in Karachi in early 2010
-
8/12/2019 PBR 1007 Complete
51/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
273
Some new investigations on the basis of the behavior of the
consumers of mobile services in this institute are being included.
(Q.2 to Q.7) in our instrument attached
In this research it is being investigated:
Whether Data services (e.g. SMS, RINGTONES, MUSIC,
and DOWN LOADS) have greater effect on perceived
economic value for students than for staff?
Whether Data services (e.g. SMS, RINGTONES, MUSIC,
DOWN LOADS) have greater effect on perceived
emotional value for staff than the students.
Whether Functional service quality (billing system) hasa greater effect on perceived value for staff than for
students.
Whether Functional service quality (customer service
quality) has a greater effect on perceived emotional
value for staff than for students.
Whether Perceived emotional value has a greater effect
on satisfaction for students than the staff.
Perceived economical value has a greater effect on
satisfaction for staff than students.
Difference of perception between staff and student about
satisfaction.
Difference of perception between staff and student
about loyalty. Different mobile services would make a significant
difference in users satisfaction.
Sample design:
The targeted population of this research consists of all
students and staff (Teaching and Non-teaching) of a Business
University.
Keeping in view the variation in the perception of the
two user groups of mobile services, Stratified random sampling
has been used independently for both groups in order to select
-
8/12/2019 PBR 1007 Complete
52/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
274
representative samples for both. The data have been collected
by convenience sampling for both student and staff.
Sample size and its allocation:
For students
There were 3777 total students registered during the fall
semester, 2009 at the university. They have been classified in the
groups BBA (H), BS, MBA (Reg.) and MBA (Ex.)
Groups No. of students
BBA(H) 1428
BS 650
MBA(Reg.) 951MBA(Ex.) 748
The size of the sample with 5% level of precision and 95%
confidence level with unknown population variance is 352
The size of the sample to be selected from each stratum has been
decided by proportional allocation method. The calculated sizes
of the samples are 132, 61, 88, and 69 respectively.
For staff
Total staff of the university in Fall 2009 was 342. This
can be split in three categories i.e. management staff (111),permanent faculty (106) and visiting faculty (125).
The size of the sample with 5% level of precision and
95% confidence level with unknown population variance is 150
The size of the sample to be selected from each category
has been decided by proportional allocation method. The
calculated sizes of the samples are 48, 42 and 60 respectively.
Samples from both the groups have collected by convenience
sampling.
-
8/12/2019 PBR 1007 Complete
53/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
275
1) Correlation analysis has been used to explore
the relationship between different concepts
and attributes
2) Regression analysis is being used to determine
direction of causation.
3) Multiple regressions are being used to develop
regression model for Loyalty.
4) Tes ting of hypothesis has been performed
by using t- stat istic, Chi- square statistic,
analysis of variance etc.
Data Analysis:
Table 1-
Correlation analysis results based on the data of the
samples for student and staff given in Table 1
F Sig. t df
Sig. (2-
tailed)
Mean
Difference
Std. Error
Difference Lower Upper
Equal variances
assumed
.747 .388 2.146 467 .032 .17321 .08070 .01464 .33179
Equal variances no
assumed
2.057 217.833 .041 .17321 .08420 .00727 .33916
Equal variances
assumed
3.027 .083 3.480 495 .001 .20277 .05826 .08829 .31724
Equal variances no
assumed
3.286 240.689 .001 .20277 .06171 .08120 .32433
billing
service
data
services
Independent Samples Test
Levene's Test for t-test for Equality of Means
95% Confidence
It is found that there is significant difference in the
perception towards the billing service between the groups of
students and staff as P
-
8/12/2019 PBR 1007 Complete
54/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
276
Table 2 Correlations between Emotional Value and Loyalty
classification of respondents emotional value loyaltyStudents emotional
value
Pearson Correlation 1 .530**
Sig. (2-tailed) .000
N 351 351
loyalty Pearson Correlation .530**
1
Sig. (2-tailed) .000
N 351 351
Staff emotional
value
Pearson Correlation 1 .677
Sig. (2-tailed) .000
N 149 147
loyalty Pearson Correlation .677**
1
Sig. (2-tailed) .000N 147 147
**. Correlation is significant at the 0.01 level (2-tailed).
H1a
stated customer loyalty will be positively influenced by customer emotional value.
The effect of customer emotional value on loyalty is significant as p
-
8/12/2019 PBR 1007 Complete
55/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
H1b
stated customer loyalty will be positively influenced by customer economical value.
The effect of economic value on loyalty is significant as p
-
8/12/2019 PBR 1007 Complete
56/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
Table 5 Correlations between Emotional Value and Satisfaction
classification of respondents emotional value Satisfaction
Students emotional
value
Pearson Correlation 1 .542
Sig. (2-tailed) .000
N 351 351
satisfaction Pearson Correlation .542 1
Sig. (2 -tailed) .000
N 351 351
Staff emotional
value
Pearson Correlation 1 .602
Sig. (2-tailed) .000
N 149 149
satisfaction Pearson Correlation .602** 1
Sig. (2 -tailed) .000N 149 149
**. Correlation is significant at the 0.01 level (2-tailed).
H2b
stated satisfaction will be positively influenced by customer emotional value. The
effect of perceived emotional value on satisfaction is significant as p
-
8/12/2019 PBR 1007 Complete
57/201
Research
PAKISTAN BUSINESS REVIEW JULY 2010
Differences in Mobile Service Perceptions
H3stated Data services will have a greater effect on perceived economic value for
students than for staff. The effect of data services on economic values is significant
for students as p
-
8/12/2019 PBR 1007 Complete
58/201
PAKISTAN BUSINESS REVIEW JULY 2010
Research Differences in Mobile Service Perceptions
Table 8a
classificaton ofrespondents Model R R Square
Adjusted RSquare
Std. Error of theEstimate
student 1 .584a .341 .329 .62607
Staff 1 .747b .558 .538 .55288
a. Predictors: (Constant), satisfaction, network quality, economic value, customer support, data services, emotional value
b. Predictors: (Constant), satisfa