Exploring Perceived Value and Usage of Information Systems in Government Context

(1)

ChiaoDa M.αnagement Review Vol. 33 No.2, 2013 pp.75 -104

政府機關資訊象統知覺價值與使用之

關條

Exploring

Perceived

Value

and

U

sage of Information

Systems in

Government

Context

許芳銘 1 Fang-Ming Hsu 國立東華大學資訊管理學象

Department ofInformation Management, National Dong Hwa University

范秋足 Chiu司TsuFan

東南科技大學企業管理學系

Department of Business Adminis仕ation， Tungnan University

陳澤義 Tser-yiethChen

國立台北大學國際企業研究所

Graduate Institute of Intemational Business, National Taipei University 王淑玟 ShuwenWang

聖母醫護管理專科學校健康與休閒管理科

Department ofHealth and Leisure Management, St. Mary's Junior College ofMedicine, Nursing and Management

摘要:政府機關與私人企業不同，因此其資訊系統之使用行為將有所差異。本研究探討政府機關中的資訊系統情境到使用者行為之間的因果關餘，透過實務調查，暸解主觀規範、知覺品質、知覺價值、滿意度與使用資訊系統意

圖之間的關條。研究結果顯示政府機關資訊系統之使用意圖顯著地受到使用

1 Corresponding author: Dep訂恤entof Infonnation Management, National Dong Hwa University, Hualien, Taiwan, E-mail: fmhsu@mail.ndhu.edu.tw

Acknowledgment: The authors would like to thank the National Archives Administration of the Republic of China for fmancially supporting this research.

(2)

76 Exploring Perceived Value and Usage of Information Systems in Government Context

者之社會價值與功能價值的影響，而不受情境價值的影響。不同於一般企業環境，政府機關資訊系統使用者的使用意圖並不受其滿意度的影響。

關鍵詞:知覺價值;主觀規範;資訊系統成功;政府

Abstract: The usage of information systems in government agencies is different to that in business. This study investigates the causal relationship of information systems (IS) usage from IS context to user's behavior in government settings. Through empirical s盯vey， the relationship among subjective norm, perceived quality

,

perceived value

,

satisfaction and intention to use IS in government was examined. Analytical results indicate that user intention to use IS in government is markedly inf1uenced by social value and functional value and not by conditional value. It is interesting to find that in government

,

in con甘ast to the business environment

,

users' intention to use IS does not count on their satisfaction.

Keywords: Perceived value, Subjective norm, IS success, Government

1. Introduction

The effect of subjective norm, quality and perceived value on behavioral intention to use information systems (IS) in government agencies was examined. Intention to use IS is a function of user's quality perceptions of the system, information and services (DeLone and McLean, 2003). Perceived value is increasingly recognized as essential to intention in research. User behavior is positively correlated with hislher perception of value (Sheth, Newman and Gross,

1991). In order to develop superior online services

,

government agencies must understand the factors and path that influence user adoption of IS (Carter and Belanger

,

2005). Recent advances in consumer and services marketing consider that perceived value and satisfaction are central to explaining customer loyalty (Fiol, Tena, and García, 2011). However, no study has characterized the relationships among IS context, users' internal value and behavioral intention to

(3)

Chiao Da Management Review Vol. 33 No.2, 2013 77

use IS in government. This study tries to make a contribution to the research area of IS usage in government through the mediating of perceived value.

Perceived value comes from performance outcome and reflects on customer behavior in IS context (H凹， 2006; Hsu, Chen, and Wang, 2010). Behavior is motivated by perceived values (Holbrook

,

1994; Shukla

,

2010). The value framework developed by Sheth

,

Newman

,

and Gross (1991) explains why consumers make choices based on their perceived values. Value and attitude affected by context and previous experience deterrnine human behavior.

Informational environment shapes the online service use and behavior (Lamb,

King and Kling, 2003). Past use is the most important predictor of intention to use and future use, and its impacts increase as past usage increases (Kim and

Malho仕a， 2005). Perception of quality has effects on desirable outcomes such as consumer intent to recommend a service. Besides, performance has an indirect effect on behavior mediated by consumer-perceived value (Venkatesh and Daw芯，

2000). Bagozzi's 企amework described self-regulating processes in terms of emotional response to behavior with various outcomes (Chang, Wang and Yang, 2009). Perceived sacrifice, perceived service quality, and perceived value 企om

the context of service environment affect the consumers' behavioral intentions (Shukla, 2010). This study proposed a context-value-behavior approach by adding perceived value to the context construct and specific behavior of satisfaction and intension to use IS in government. This study tries to underline the relationship in which contexts affect perceived value and subsequent behavior.

The intention to use IS in government wiU be different than that in business. Value within business markets is often perceived as the moneta可 price or net worth of the costs and benefits that are associated with a particular exchange

(Stewa肘， 2006). However money or profit is not the major concem in government.

Electronic government is no longer merely an option but has become a necessity (Gupta and Jar油， 2003). In the e-government context, government agencies have to use IS. Based on the s仙diesby Shannon and Weaver (1949) and Mason (1978),

DeLone and McLean (1992; 2003) proposed a model of IS success to measure whether IS are effective in organizations. In this model

,

IS system quality and information quality influence use status and user satisfac

(4)

78 Exp/oring Perceived Va/ue and Usage of Information Systems in Government Context use status and user satisfaction then influence user's behavior

,

and

,

in turn

,

affect organizational perforrnance (DeLone and McLean, 1992). User value is an antecedent judgment of usage. Therefore

,

the antecedents of usage can then be further deconstructed into different perceived values. They are mediating effects between context and user's behavior. Once the context-value-behavior 企amework is approved by government setting, the relationship which contextual factors influence users' perceived values and values influence user's behavior will be understood.

2. Literature

Review and Hypotheses

From the perspective of technology accep個nce model (T AM), subjective norrn influences behavioral intention (Venkatesh and Davis, 2000). Institutional theory argues that stakeholders impose subjective norrns on frrms

,

including governrnent agencies (Delmas and Toffel

,

2004). Subjective norrn is based on political support for stringent regulations and regulatory threats that impede operations based on operational performance. Subjective norm is “the realization of those important reference people to a person's perception of whether he/she shall carry out such a behavior" (Fishbein and Ajzen

,

1975). Even ifpeople do not favor perforrning a given behavior, if they believe that important reference person believe that they should perforrn this behavior, then they willlikely perforrn such a behavior (Ho and Yang, 2009). Due to the characteristics of governrnent, regulations based on rules and laws are important to IS operations in governrnent agencies. Therefore

,

subjective norrn that regulates behavior of governrnent IS users is included in this study.

The IS success model proposed by DeLone and McLean (2003) encompasses IS quality characteristics (system quality)

,

quality of IS processing (service quality), quality of IS output (inforrnation quality), consumption of IS output (usage), user reaction to the IS (user satisfaction), and IS influence on user behavior (intention to use). User satisfaction is a user reaction to the using of IS output. Service quality is closely related to user satisfaction, whereas satisfaction

(5)

is defined as the gap between expectation and feelings toward the service rendered (Pi哎， Watson, and Kavan, 1995). Thus, system quality, information quality and service quality are determinants of user's intention to use IS (DeLone and McLean

,

2003).

Perceived value comes from performance outcome and ref1ects user's behavior. Perceived value is a construct representing how consumers judge

“

what was received to acquisition costs (e.g., fmancial, psychological, effort)" (Oliver,

1997). Zeithaml (1988) defines value as

“

the consumer's overall assessment of the utility of a product based on perceptions of what is received and what is given." Perceived value is a customer's perceived preference for and assessment of product attributes, attributes performance, and consequences arising from use that facilitate achieving customer objectives and purposes in use situations (Woodruff

,

1997). This defmition links the perceived value of customer to product or service use

,

and attributes perceived value to the perceived difference between what a consumer perceives and what they pay in exchange for what they receive (Woodruff

,

1997). Behavioral intentions represent various consumer responses such as intention to use a product or service. Perceived quality is a consumer's perception of

“

the totality of features and characteristics of a product or service that bear on its ability to satis命 stated or implied requirements" (Oliver,

1997). Service quality is c10sely correlated with user satisfaction

,

in which satisfaction is defmed as the gap between expectations and the perception of service rendered (Pi役， Watson, and Kavan, 1995). Furthermore, system quality and information quality inf1uence user satisfaction and IS usage. User satisfaction and IS usage ref1ect individual behavior (DeLone and McLean, 1992). Thus，企om

the TAM and IS success model, this study takes subjective norm, system quality,

information quality and service quality as antecedents of perceived value to examine user satisfaction and intention to use IS in government.

Sheth, Newman and Gross (1991) explained why consumers make choices based on consumption value and identifies the following five consumption values that inf1uence consumer behavior: functional, social, emotional, epistemic, and conditional. Sweeney and Soutar (2001) identified the value construct and measures customers use when analyzing a product prior t

(6)

measures can be applied to assess customer perceptions of the value of consumer durable goods at a brand level within a purchase context and to determine which consumption value underscores purchase attitudes and behaviors. As government is a relatively stable entity

,

epistemic value and emotional value are not appropriate in government settings. The inf1uence of epistemic value on commitment is not significant (Pura, 2005). Therefore, this study uses conditional value, functional value and social value as constructs of value in analyzing the intention to use IS in a government context.

Several hypotheses were proposed in this study. Subjective norm means that a potential decision maker recognizes a need for making decisions and is inf1uenced by individualism (Hartwick and Barki, 1994). Since subjective norm means that when an individual recognizes that he is a social actor, he will be expected to play an appropriate role by society and thus inf1uenced by others. Consumers with a high degree of value congruence within an organization appear to be more willing to share information (Caz肘， Shao, and St Louis, 2007). Social value comes from the association of the product with the social group (Pope, 1998) and is the perceived utility obtained 企om an altemative's association with one or more particular social groups (Sheth, Newman and Gross, 1991). Social value enforces particular behaviors and rewards or punishes these behaviors (Sweeney, Soutar, and Johnson, 1999). Subjective norm is positively associated with social value. Furthermore

,

Holbrook (1994) argued that social value was related to one or several social groups based on previous or present geographic locations. Interactions among the economy, culture, morality, and groups produce social value and specific symbolic meanings that satisfy self-valued performance. Value of an excellent governmental leader will have a significant effect on surrounding people and form the extent of social value. Subjective norm positively affects the social value of customers (H凹， Chen, and Wang, 2010). Since operations of electronic records management systems (ERMS) are regulated by the Archives Act in Taiwan

,

government agencies are inf1uenced by the requirements of upper-level ones or regulating agency called the National Archives Administration (NAA). Thus, we propose the following hypothesis.

(7)

Chiao Da Management Review Vol. 33 No.2, 2013 81

Hl: Subjective norm is positively related to social value.

Infonnation quality, namely the quality of infonnation offered by systems,

inc1udes measures of IS outputs primari1y in the fonn of reports (DeLone and McLean, 1992). A semantic instrument for infonnation quality had measures of relevance

,

accuracy

,

timeliness

,

aggregation

,

and fonnatting (Ahituv

,

1980). The functional value of an altemative represents the utility it is perceived to possess on criteria salient to its utilitarian or physical purposes (Sheth, Newman and Gross

,

1991). Infonnation quality positively affects perceived value in the electronic context (Wang

,

2008). The relevance of infonnation to user's task or problem will render the perceived functional value. The matches between outputs of an IS and user's requirement affect the subjective value of infonnation

,

inc1uding assignment relevance, richness of infonnation, and immediateness. Additional旬， when users recognize that infonnation meets their needs, this

recognition fonns a functional value of this infonnation. Therefore, infonnation quality positively affects functional value. Moreover

,

user recognition of quality for an IS is positively correlated with functional value. Restated

,

infonnation quality is positively correlated with functional value. Therefore

,

we propose the following hypothesis.

H2: Information quality is positively related to j切1.ctionalvalue.

In IS context, service quality is essential to its success (Kettinger and Lee,

1995; Pitt

,

Watson

,

and Kavan

,

1995). Service quality sourced 企om IS success model indicates that an infonnation provider offers various services for infonnation recipients

,

inc1uding time1y service

,

consideration

,

and assurance (DeLone and McLean, 2003). Value is an interactive experience (Woodruff,

1997). Service quality positively affects perceived value in the electronic context (Wang, 2008). Functional value comes from the effectiveness, practicality or utility of the product or service (Pope

,

1998). Service perfonnance directly affects customer value (Venkatesh and Dawis, 2000). Service quali可 significantly influenced functional dimension of perceived value (Chen and Hu, 2010; Wang,

(8)

the functional value concept of information recipients. Hence, we propose the following hypothesis.

H3: Service quali，砂 ispositively related to functional value.

System quality comes from the information processing system itself (DeLone and McLean, 1992). User interface, response time, reliability, and system flexibility are parts of a scheme for measuring system quality (Hamilton and Chervany, 1981). System quality positively affects perceived value in the electronic context (Wang, 2008). Conditional value is the perceived utility obtained by an altemative consumer resulting from a specific situation or set of circumstances a choice maker faces (Sheth, Newman and Gross, 1991). Conditional value comes 企om the worth attached to when or how customers use the product (Pope, 1998). An altemative's utility often depends on the situation (Sheth, Newman and Gross, 1991). Holbrook (1994) proposed that conditional value depends on the context in which a value judgment occurs, and is bound to a particular situation. Since a user interface encourages users to use various interaction models and attain self-achievement and satisfaction, consumers are affected by system context. Moreover, reliability has a direct effect on customer value (Venkatesh and D側的， 2000). The reliability of a stable IS helps users irnprove their intelligence and skills and develop new interactive models between humans and machines. System quality is not only functions of an 路， but also generates symbolic meaning for information users. Imperceptibly, system quality is closely related to user-perceived conditional value. Hence, we propose the following hypothesis.

H4: System quali.砂 ispositively related to conditional value.

Specific behavior would be enforced by social value (Sweeney, Soutar, and Johnson, 1999). Social value is expected to have great influence on the use of

servlces ，的 theyare often used in a social environment that involves interpersonal influence (Sullivan Mort and Drennan, 2005). People's psychogenic needs are based on psychological tensions and influenced by social relations and the like. Social satisfaction comes 企om user's symbolic self-satisfaction and self-hint

(9)

consumption. U ser satisfaction has been suggested as a success measure for empirical IS researches (Hamilton and Chervany, 1981). IS user satisfaction includes measures of recipient response to the use of the output of IS (DeLone and McLean

,

1992). Perceived value positively relates to customers' satisfaction (Wang

,

2008). Social value has critical influence on customer's satisfaction (Fiol

et al., 2009). Owing to IS usage being recognized as worthy of the user's social identity and status by others

,

increase in user's perceived social value will enhance his/her satisfaction. Hence

,

H5a: Social value is positively related to user satisfaction.

User's willingness about IS reflects on his/her intention to use system (Venkatesh and Davis

,

2000). Social value is important in human relationship (Fiol, Tena, and Garc峙， 2011). The phenomenon of reference group influence is clearly germane to consumer behavior. Perceived value has positive impacts on behavioral intention (Hu, Kandampully, and Juwaheer, 2009; Hume and Sullivan Mort, 2010). IS users' perceived value positively relates to their intention to use (Hung, Tsang, and L凹， 2010). Social value is positively correlated with the intention to use 的， in which compulsory norms afford user legitimacy in using the system. Users in this situation can obtain a self-motivation to use the system (Thaler, 1985). Perceived value positively affects user's intention to reuse IS (Wang

,

2008). With regard to consumer durables

,

an ERMS user in a government agency may probably spend more effort in accomplishing records management in a computer system to appeal to his/her colleagues and the upper classes in general. Since social value has a kind of symbolic meaning of perceived value that comes 企omthe benefit for the user in using the IS

,

it will affect the user's intention to use IS positively. There is a positive relationship between social value and intentions (Pihlström and Bru品， 2008)

hypothesis.

H5b: Social value is positively related to user's intention to use IS.

Functional value refers to the satisfaction derived from the physical performing of its functions of product (Sh呦， Newman and Gross, 1991).

(10)

Perception of benefit occurs as a result of the relevance of technology to user's specific needs. The convenience of 芯， as functional value, also enhances user satisfaction. Economic satisfaction is to satisfy specific needs, solve specific problems, and produce user satisfaction. Additionally, the high-level affective value associated with an information product will support the emotional and subjective needs of a user

,

thereby generating user satisfaction. This satisfaction is a kind of non-economic psychological satisfaction (Geyskens

,

Steenkamp, and Kumar

,

1999). Quality of IS is c10sely associated with the satisfaction of users

(Pi鈍， Watson, and Kavan, 1995). It is not all for use functions of the 路， but for the IS to bring a symbolic meaning to information users. Perception of functional value brings the satisfaction of using IS in government. Functional value contributes to generate customer satisfaction (Deng et al.

,

2010). Hence

,

H6a: Functional value is positively related to user satisfaction.

Effective task fulfillment 街ives the perceived functional value which relates to superiority compared with altematives (Sheth, Newman and Gross, 1991; Pura, 2005). Perceived value positively affects user's intention to reuse 1S (Wang, 2008). Functional value that generally relates to such attributes as performance will inf1uence user's behavior and intention to use IS. Utility theory posits that reacting behavior will be the one that maXIIDlZes utility (Sheth, Newman and Gross, 1991). Users like to use the system with functions th剖 they

can easily manipulate. The professionalism of IS department inf1uences the individual IT adoption in an organization. Functional value

,

defined as utility

derived 企om the perceived quali句r and expected performance of the product or service, has a direct effect on behavioral intention (Wang et al., 2004). Functional value positively affects user's intention to use IS (Hsu, Chen, and Wang, 2010). Hence, we propose the following hypothesis.

H6b: Functional value is positively related 10 user's inlenlion 10 use IS.

Conditional value denotes the value derived 企om the independence of time and place and is experienced only in certain contexts or situations (Holbrook,

(11)

Chiao Da Management Review Vol. 33 No.2, 2013 ₈₅

1994). When a choice is driven by conditional value, the outcome is contingent on antecedent circumstances that may cause the consumers to deviate 企om their typical or planned behavior (Sheth, Newman and Gross, 1991). Perceived value positively affects user's satisfaction to IS (Wang

,

2008). The perception of users compares the difference between expected compensation and input costs (Churchill and Surprenant, 1982). The conditions that might cause users to switch IS usage

,

i.e. conditional value

,

wi11 support the subjective needs of a user. Ubiquity condition increases importance of IS services (Pura

,

2005; Nysveen

,

Pedersen and Thorbjomsen, 2005). Therefore, conditional value affects a user's satisfaction. Hence

,

H7a: Conditional value is positively related to user satisfaction.

Perceived value positively affects user's intention to reuse IS (Wang

,

2008). Conditional value often inf1uences the choice makers to deviate 企om their pattems of behavior. Conditional value refers to circumstances which impact choice and intention (Sheth, Newman and Gross, 1991). In govemment agencies, traditional records management neglected the core activity and was ignored for a long time. A good image of personnel using IS as a capable one in govemment agencies can greatly enhance the intention of records staff to use ERMS. Thus,

conditional value affects behavioral intentions of customers. Hence, we propose the following hypothesis.

H7b: Conditional value is positively related to user's intention to use IS. Figure 1 illustrates the research framework ofthis study.

3. Measurement and Data Collection

When preparing or delivering a service

,

a govemment agency typically references pertinent records or documents and, in most cases, creates new records that ref1ect and document the service rendered. Management of such documents, records, and archives is essential and tedious，企equent1y challenging govemment agencies that

(12)

86 Exploring Perceived Value and Usage of

Information Systems in Government Context

essential1y become information businesses. Several trends have emerged in the management of government records/archives, such as a

e AU O 1M eEl ra HUM--•• ohuUH --E 且‘ •• Fh e hu T

Context ~ Perceived Value • Usage Behavior

fundamental shift 企om paper-based storage to computer-based systems，台om

paper to electronic documents，企om managing information to supporting its access and retrieval, and from a cost-reduction focus to continued process improvement (Stephens, 1998). These trends all point to the need of government agencies to utilize ERMS for processing official records and archives. According to the global e-government report investigated by the Center for Public Policy of Brown University, Taiwanese e-government ranks 1st, 1st, and 2nd in 2004, 2005, and 2006 respectively (West, 2006). Therefore, this study uses an ERMS as the target IS in e-government. From the Archive Act, each government agency in Taiwan is responsible for managing its official records electronical1y with

(13)

necessary accessibility and security. When issuing or recelvmg an official document, an agency must create electronic records and transfer record catalogs to the NAA through online systems developed by NAA. This study takes the government as an arena of consumption in which employees engage in quests for perceived value, and construct and confrrm their identities as consumers of IS.

Through the proposed model

,

specific constructs were examined and then operationalized using relevant measures obtained 企om previous researches. Particularly, this study evaluates system quality, information quality, service quality, subjective norm, user satisfaction and intention using items adopted 企om studies by DeLone and McLean (2003) and Venkatesh and Davis (2000). In government, IS users must follow rules and laws instead of epistemic value. Moreover, users in government must use IS rationally without attaching emotional value to the IS. Perceived value is measured by items adopted from the work of Sheth, Newman and Gross (1991), i.e., conditional, functional, and social value constructs. These items pertain to cornmon problems regarding user value and ERMS usage in government.

All questionnaire items were obtained from previous studies and modified for this study (Sheth, Newman and Gross, 1991; Venkatesh and Davis, 2000). In total, 19 indicators for nine constructs in this study were assessed using a modified 7-point Likert-type scale without the middle point

,

ranging 企om 6 for the most positive evaluation and 1 as the most negative evaluation. The performance criterion was adopted by concept of TAM proposed by Venkatesh and Davis (2000). Additional旬， perceived value is measured by items adopted 企om the work of Sheth, Newman and Gross (1991), i.e., social, functional, and conditional value constructs. In this paper, we employ the perceived value and technology acceptance standard to conduct evaluation of IS user's behavior. 1) Subjective norm

,

containing two indicators: obligation and compliance. 2) Information quality

,

containing two indicators: importance and relevance of reports. 3) Service quality, containing two indicators: benefit and demons仕ation of help. 4) System quality, containing two indicators: reliability and 企iendliness. 5) Social value, containing two indicators: regulation fulfillment and supervision attainment. 6) Functional value, containing three indicators

(14)

88 Exp/oring Perceived Va/ue and Usage of lnformation Systems in Government Context

report sufficiency, and operation efficiency. 7) Conditional value, containing two indicators: prestige and capability perception. 8) User satisfaction, containing two indicators: satisfaction of ERMS and auditing system. 的 Intention to U se, containing two indicators: usage 企equency and reuse tendency. These items pertain to common problems regarding user's perception of value and ERMS usage in government context.

Five domain expe此s reviewed a preliminary questionnaire and provided analytical feedback. These experts were IS managers in government agencies and NAA managers highly knowledgeable regarding elec甘onic record management practices. Based on assessments and suggestions, several changes were made to item wording, fine-tuning the language to reach out to the target agencies and employees. Survey layout was redesigned to improve its visual appea1. A pilot study was then conducted to assess the survey instrument using key personnel from 10 government agencies. The top three levels of cen仕al governments, i.e. Yuan, minis甘y and bureau level, and top two levels of local governments, i.e. county and town level, were inc1uded as sample.

Questionnaire packets were mailed to 1,700 government agencies. The survey packet consisted of a cover letter describing study objectives and the da個 management plan, a support letter from the NAA, and the questionnaire. Completed questionnaires with signatures of chief officers in an agency were collected t尬。ugh the official reporting channel and website. 360 central agencies completed the survey, for an effective response rate of 21.8%. The survey targeted s個ffs in charge of managing electronic records who understand the implementation and current practice ofERMS in their agency.

4. Modeling Results and Discussion

4.1

Initial Model

Construct validity is assessed by examining the evidence associated with each construct, inc1uding appropriate items with loading of minimum 0.4 on their

(15)

respective hypothesized components in confirmatory factor analysis (CFA)

(Anderson and Gerbing, 1988). The initial model yields the following fit statistics: comparative fit index (CFI), 0.97; normed fit index (NFI), 0.96; non-normed fit index cl'亦可FI) ， 0.96; goodness-of-fit index (GFI), 0.88; and adjusted goodness-of-fit index (AGFI), 0.84. Although goodness-of-fit is not perfect, at 0.9 it is still within the acceptable range (JOreslog and SOrbom, 1984). Root mean square error of approximation (RMSEA) is 0.079 ançlx2/ df= 3.27, indicating

that the hypothesized model fits data reasonably wel1. The situations, GFI and AGFI are greater than 0.8 and RMSEA is less than 0.08, make the reasonable fitness of research model.

The loadings on hypothesized factors are si伊ificant and substantial (all factor loadings exceeding 0.5). The assessment of convergent validity is supported by the CF A mode1. All loadings in Table 1 are significant (p < 0.05). Thus, convergent validity holds. Additional旬， this study also computes average variance extracted (A VE) to confirm discriminate validity (Fomell and Larcker,

1981). When the A VE value is larger than the square phi-correlation, the questionnaire has the high discriminate validi可 (Burton， Lichtenstein, Netemeyer,

and Garretson, 1998; Batra and Sinha, 2000). The AVE values are as follows: subjective norm, 0.80; information quali旬， 0.92; service quality, 0.87; system quality, 0.50; social value, 0.41;

“

nctional value, 0.37; and conditional value,

0.87. Moreover, the measure of user satisfaction is 0.59 and intention to use is 0.81. The A VE values for every variable demonstrate that this study has discriminate validity.

The quali句， of measurement efforts by investigating reliability, convergent validity, discriminate validity and construct validity were also determined. This study assesses reliability for all items in a construct by calculating composite reliability (CR). All composite reliability values in this study are larger than 0.50, indicating an acceptable fit to data (Fomell and Larcker 1981). This result shows that this study has good reliability. The intemal consistency reliabilities are as follows: subjective norm, 0.89; information quality, 0.96; service quality, 0.93; system quality, 0.67; social value, 0.58; 如lctional value, 0.64; conditional value,

(16)

90 Exploring Perceived Va/ue and Usage of Information Systems in Government Cont，臼t

Table 1

Confirmatory Factor Analysis Construct and Measurement Items

Subjective Norm

People who inf1uence my behavior think that 1 should use ERMS People who are important to me think that 1 should use ERMS Information Quality

The output reports ofERMS are important to my job The output information of ERMS is relevant to my job Service Quality

IS staffhelps me realize the benefit ofERMS IS staff helps me demons甘'atethe results of ERMS System Quality

TheER此1Ssystem is reliable The interface of ERMS is 企iendly

Social Value

Using ERMS fulfills the requirements ofthe National Archives Using ERMS fulfills the requirements ofupper-level government Functional Value

ERMS can present the contents of records and attachments ERMS can provide reports for electronic records management ERMS can improve the operation efficiency of electronic records Conditional Value Loadings 0.88 0.91 0.94 0.98 0.93 0.94 0.64 0.77 0.59 0.68 0.52 0.62 0.68 Getting high prestige 企omothers would cause me to use ERMS 0.92 Regarding as a capable user would cause me to use ERMS 0.95 User satisfaction

1 am satisfied with ERMS 0.64

1 am satisfied with auditing system for electronic records 0.88 Intention to Use

1 plan to use ERMS in the next month 0.90

Assuming 1 can access to ERMS, 1 intend to use it 0.90 The hypotheses testing result of the given model in Table 2 indicates that most relationships among latent constructs significantly support the hypotheses, thus providing initial evidence for the conceptual models in this study and supporting mono-logical construct validity. Subjective norm is an antecedent of social value (H1: y=0.38). Additionally, information quality and service quality significantly affect functional value (H2， γ=0.19; H3, y=0.82). System quality positively influences conditional value (H4:γ=0.41). Social value positively influences user satisfaction and intention to use (H5a，戶=0.13; H5b

,

ß=0.35).

(17)

Functional value positively inf1uences user satisfaction and intention to use (H6a

,

戶=0 .48; H6b，戶 53=0.67). Conditional value significantly affects user satisfaction (H7a: ß=0.14) and does not si伊ificantlyaffect intention to use (H7b: ß=0.03).

:' Table 2

Results of Hypotheses Testing of the Initial Model Estimated

Hypothesized Path Coefficient t-value

H1: Subjective norm 今 SocialValue 0.38 4.82料

H2 : Information Quality-多 Functional

0.19 4.00**

Value

H3: Service Quality 今 FunctionalValue 0.82

&:

System Quality 今 ConditionalValue 0.41 HSa: Social Value 今 Usersatisfaction 0.13 HSb : Social Value 今 Intentionto Use 0.35 Hω: Functional Value 今 Usersatisfaction 0.48

H的: Functional Value 今 Intentionto Use 0.67 H7a : Conditional Value 今 User satisfaction 0.14

H7b : Conditional Value 今 Intentionto Use 0.03

-dd t-ee d 一此前內一 oo m-PP 3uw-nynr E-uu -nδσδ 11.36** Supported 7 .1 5 抖 _Supported 1.91 * Supported 5 .44料 _Supported 5.77** Supported 10.66** Supported 2.36** Supported 0.70 Not Significant Note: *: p < 0.05, **: p < 0.01

Subjective norm

,

information quality

,

service quality

,

and system quality positively and significantly inf1uence the perceived value of users. Users' perceived value affects their levels of satisfaction significantly. Intension to use is impacted by functional and social value rather than conditional value. Users' satisfaction does not affect their intention to use IS in government settings. The most crucial contextual factor is service quality, which affects functional value and, therefore, strongly impacts user satisfaction and intention to use.

4.2

Rival Model

The service marketing literature indicates that perceived satisfaction strongly impacts consumer intention . Behavioral intentions are strongly

(18)

92 Exp/oring Perceived Va/ue and Usage of Information Systems in Government Context

inf1uenced by satisfaction with office-related performance factors in a retail context (Van Birgelen, De Jong, and De Ruyter, 2006). Research investigating consumer satisfaction-behavioral consequence frameworks produced a

substantive body of evidence regarding the direct effects of customer satisfaction on behavioral intentions (Cronin, Brady, and Hult, 2000; Seiders et al., 2005).

The theoretical rationale for the model，企equently left implicit in the literature,

can be derived using Bagozzi's (1992) appraisal-emotional response-coping

f泊mework. His 企amework proposes that customer perceptions (i.e., appraisals) trigger customer satisfaction (i.e., affective response), which in turn stimulates the development of future favorable behavioral intentions

,

often directed at maintaining satisfaction levels. Moreover

,

given the relatively high degree of perceived risk associated with services, post-evaluation of an actual experience is transformed into an important antecedent of consecutive behavior (Bennett, Hartel,

and McColl-Kennedy, 2005). Via increased value, a

‘

lock,..in' effect can occur

through which satisfied customer will intend to use an IS for future 仕ansactions.

Hence, hypothesis H8 -User satisfaction positively affects user's intention to use IS - is added in the rival model.

The rival model produces the following fit statistics: CFI, 0.97; NFI, 0.96; NNFI, 0.96; GFI, 0.88; and AGFI, 0.84. The RMSEA is 0.080 ançlx2 / df= 3.29, implying that the rival model fits the data well (Figure 2). The altemative rival model suggests no direct path exists between satisfaction and intention to use. The hypothesized model is compared with the rival model using overall 缸， GFI， AGFI，

CFI, and NFI for each model' s parameters that are statistically significant. The criteria of the rival model are inferior to those of initial model. Additionally

,

the ratio of the number of significant paths in the rival model (9 of 11) is lower than that of the proposed model (9 of 10).

(19)

Figure 2 The Rival Model

4.3

Discussion

In addition to the constructs of perceived quality, satisfaction and behavioral intention, perceived values were added to detennine the relationship between perceived quality and subjective nonn along with perceived value to identify the direct and indirect effects on consumer reactions to satisfaction and intention to use in a government context. This study identifies the important at仕ibutes/cues in governmental IS settings, which may be utilized to review characteristics of IS context and usage as experienced by governmental users. The context-value-behavior model of this study for IS in government is adapted 仕om the IS success model developed by DeLone and McLean (2003) and consumption value proposed by Sheth, Newman, and Gross (1991).

Perceived value indeed acts as the mediating effect between context outcomes and IS user's behavior 企ominvestigated findings. Based on the concept of positive reinforcement, behavior is affected by outcomes (Robbins, 2003).

(20)

94 Exp/oring Perceived Va/ue and Usage of Information Systems in Government Context Additionally, performance outcome is the former concept of customer value (H凹， 2006). Therefore

,

the customer value inf1uences IS users' behaviors. From the analytical results, user satisfaction to IS in government is affected by conditional value

,

functional value

,

and social value. The intention to use IS is affected by functional value and social value rather than conditional value. The proposed modified model of the IS success in government context with subjective norm and value factors added is more elaborate than the original one, suggesting that in the government context, the context-value-behavior model proposed by this study is acceptable for IS in government. Notably, research findings suggest that intention in government is significantly affected by functional value and subjective norm rather than satisfaction 企om the comparison between initial model and rival model proposed by this study. This result differs from that obtained by Shank: Venk:atesh and Dawis (2000) in a business environment which argues satisfaction has a direct impact on consumer's intent to do some things

,

implying that in the government context the role of satisfaction differs 企om the conventional business context of user behavior.

User satisfaction and intention to use IS in government is affected primarily 企om functional value in terms of information quality and service quality. This finding is similar to that obtained by Sheth et al. (1991)

,

who argues that economically rational behavior is generally analogous to behavior motivated by functional value. This finding is similar to those obtained by other studies in that information quality has a stronger relationship with customer satisfaction than the system quality itself (H凹， 2006). Compared with the coefficients of the paths in the research model

,

service quality is the most important factor which strongly inf1uences functional value and thereby significantly inf1uences user satisfaction and intention to use IS. Executives of the agencies could improve the service quality to IS users for high functional value, user satisfaction, and intention to use m government.

Consumer intentions can be predicted 企om attitudes that can further be predicted using subjective norm based on the the的 of planned behavior.

Subjective norm is higher in government settings than that in business contexts. The intention to use IS is affected significantly by subjective norm through the

(21)

mediating of social value. Although system quality significantly affects

conditional value, conditional value does not affect intention to use 1S in government. Governmental users have to use 1S under subjective norm in spite of their conditional value measured by prestige and capability image. However, Venkatesh and Davis (2000) considered an extension of technology acceptance model which focuses on antecedents of perceived usefulness, inc1uding concepts of social influence process (such as subjective norm and image) and cognitive instrurnental process. Since image is important to users' perception of the social context and atti仙des to 1S, image could be the driving force behind consumer intentions to use 1S.

5. Concluding Remarks

The most typical assumption of research in consumers' behavior is that they are voluntary (Brown and Venkate品， 2005). Prior 1S researches have been conducted primarily in environments of voluntary adoption, the applicability of previous findings to the mandatory usage context is unc1ear (Chan et al., 2010).

For 1S usage in government, a context-value-behavior approach is proposed. Perceived value plays a mediating role between context and users' behavior. The ability to form representations of perceived value that can be flexibly activated, depending upon context, is a fundamental aspect of human intention and behavior. It's important to note that context can be established and maintained by government officers and information staff. By combining the 1S success model and consumption value approaches, we hope to shed light on the mechanisms under1ying a form of positive reinforcement in which contexts dictate how perceived value predicts behavior. Our work might also lead to more general principles that can advance understanding of perceived value for which context is important. 1n particular, our 企amework explains how context, when combined with users' perceived value, can affect users' behavior in a variety of government agencles.

(22)

96 Exploring Perceived Value and Usage of Information Systems in Govemment Context

The intention to use IS is affected by functional value and social value rather than conditional value. The functional value has the strongest influence on users' satisfaction and intention. Since users in mandatory setting have to fulfill the regulation first, they expect that IS can attain their task requirements. Therefore

,

the deployment of IS has to focus on the functionality than social and conditional value. More contextual factors, e.g. relationship network, influencing users' social value and functional value could be added into fu個re research such that influence path 企om operating context through perceived value to behavior wi11 be well understood. In terms of path coefficients

,

this study encourages the strong support of IS staff to users for better service quality to IS success in government owing to their lack of IS proficiency.

Subjective norm is recommended to be added as an important dimension of IS success in the e-government environment where regulations is crucia1. In contrast with the business environment

,

it is interesting to find that users' intention to use IS does not count on their satisfaction in government settings.

From e-government policy, all agencies must provide catalogs oftheir records and

archives with a pre-specified data format through on-line. delivery to NAA periodically. Subjective norm is higher in government settings than that in business context. Regulatory compliance is often the reason for intensive IS use in government. Comparing the initial model with the rival model of this study

,

the intention to use IS is not significantly affected by user satisfaction in government

context under the existence of subjective norm or regulation. Therefore,

subjective norm is a stronger factor in the government context than in business envlronment.

In the aggregate

,

this 企amework depicts detailed relationship and projects better estimates of IS success in government. Conceming specific contributions

,

findings of this study could help managers tailor their computerization context to enhance users' satisfaction and intention to use IS in government agencies. Future studies could incorporate a longitudinal measurement to address the dynamics of perceived value and behavior in IS acceptance.

(23)

Chiao Da Management Review Vol. 33 No.2, 2013 ₉₇

References

Ahituv, N. (1980),“A Systematic Approach toward Assessing the Value of an Information System," MIS Quarter鈔， 4(4)， 61-75.

Anderson, J. C. and Gerbing, D. W. (1988),“Structural Equation Modeling in Practice: A Review and Recommends Two-step Approach," Psychological Bulletin, 103(3),411-423.

Bagozzi, R. P. (1992), “The Self-regulation of Attitudes, Intentions and Behavior," Social Psychology Quarterly, 55(2), 178-204.

Batra, R. and Sinha, 1. (2000),“Consumer-Level Factors Moderating the Success ofPrivate Label Brands," Journal of Retailing, 76(2), 175-191.

Bennett, R., Hartel, C. E. 1. and McColl-Kennedy, J. R. (2005),“Experience as a Moderator of Involvement and Satisfaction on Brand Loyalty in a Business to Business Setting," lndustrial Marketing Management, 34( 1), 97-107.

Brown, S. A. and Venkatesh, V. (2005),“Model of Adoption of Technology in Households: A Baseline Model Test and Extension Incorporating Household Life Cycle," MIS Quarter鈔， 29(3), 399-426.

Burton, S., Lichtenstein, D. R., Netemeyer, R. G. and Garretson, J. A. (1998),“A Scale for Measuring Attitude toward Private Label Products and an Examination of Its Psychological and Behavioral Correlates," Journal of the Academy of Marketing Science, 26(4), 293-306.

Carter, L. and Belanger, F. (2005),“The Utilization of E-government Services: Citizen Trust, Innovation and Acceptance Factors," lnformation System Journal, 15(1), 5-25.

(24)

98 Exp/oring Perceived Va/ue and Usage of

Cazier, J. A., Shao, B. B. M. and St Louis, R. D., (2007),“Sharing Information and Building Trust through Value Congruence," Information Systems Frontiers

,

9(5)

,

515-529.

Chan, F. K. Y., Thong, 1. Y. L., Venkatesh, V., Brown, S. A., Hu, P. J. H. and Tam, K. Y. (2010),“Modeling Citizen Satisfaction with Mandatory Adoption of an E-Government Technology," Journal of the Association for Information Systems

,

11(10)

,

519-549.

Chang, H. 旺， Wang, Y. H. and Yang, W. Y., (2009),“The Impact of E-service Quality, Customer Satisfaction and Loyalty on E-marketing: Moderating Effect ofPerceived Value," Total Quality Management, 20(4), 423-443. Chen, P. T. and Hu, H. H., (2010),“How Determinant Attributes of Service

Quality Influence Customer-perceived Value: An Empirical Investigation of the Aus仕alian Coffee Outlet Industry," International Journal of Contemporary Hospitality Manageme肘， 22(4), 535-551.

Churchi11, G. A., Jr. and Surprenant, C. (1982),“An Investigation into the Determinants of Customer Satisfaction

,"

Journal of Marketing Research

,

19(4),491-504.

Cronin, J. J., Brady, M. K. and Hult, G. T. M. (2000),“Assessing the Effects of Quality, Value and Customer Satisfaction on Consumer Behavioral Intentions in Service Environments," Journal of Re的iling， 76(2), 193-218.

Delmas, M. and Toffel, M. W. (2004), “Stakeholders and Environmental Management Practices: An Institutional Framework," Business Strategy and the Environment

,

13(4)

,

209-222.

DeLone, W. H. and McLean, E. R. (1992),“Information Systems Success: The Quest for the Dependent Variable," Information Systems Research, 3(1),

(25)

DeLone, W. H. and McLean, E. R. (2003),“The DeLone and McLean Model of

Information Systems Success: A Ten-Year Update," Journal of Management

Information 砂'stems， 19(4),9-30.

Deng, Z., Lu, Y., Wei, K. K. and Zhang, J., (2010),“Understanding Customer Satisfaction and Loyalty: An Empirical Study of Mobile Instant Messages in China," International Journal of Information Management, 30(4), 289-300. Fiol, L. 1. c., Alcañiz, E. B., Tena，已tl. A. 孔tl. and García, J. 鼠， (2009),“Customer

Loyalty in Clusters: Perceived Value and Satisfaction as Antecedents," Journal ofBusiness to Business Marketing

,

16(3)

,

276-316.

Fiol, L. J. 仁， Tena, M. A. M. and Garc泊， J. S. (2011), “Multidimensional Perspective of Perceived Value in Indus仕ial Clusters," The Journal of Business & lndustrial Marketing

,

26(2)

,

132-145.

Fishbe血， M. and Ajzen, 1. (1975), Bel呦， Attitude, lntention and Behavior: An lntroduction to Theoη'and Research

,

Reading

,

MA: Addison-Wesley.

Fomell, C. and Larcker, D. F. (1981),“Evaluating Structural Equation Models with Unobservable Variables and Measurement Error," Journal of Marketing

Reseαrch， 18(1),39-50.

Geyskens, 1., Steenkamp, J. E. M., Kum缸， N. (1999), “A Meta-analysis of Satisfaction in Marketing Channel Relationships," Journal of Marketing Research, 36(2), 223-238.

Gupta, M. P. and Jana, D. (2003),“E-govemment Evaluation: A Framework and Case S仙dy，"Government lnformation Quarter鈔，20(4)， 365-387.

Hamilton, S. and Chervany, N. L. (1981), “Evaluating Information Systems Effectiveness: Part 1 Comparing Evaluation Approaches," MIS Quarter紗， 5(3)，

(26)

100 Exploring Perceived Value and Usage of

lnformation Systems in Government Context

Hartwick, J. and Barki, H. (1994),“Explaining th Role of User Participation in

Infonnation System Use," Management Science，的(4)，440-465.

Ho, S. H. and Yang, T. Y. (2009),“Understanding Consumer Acceptance of

Intemet Banking-From Stayers and Switchers' Perspectives," Chiao Da

Management Review

,

29(2)

,

1-20.

Holbrook, M. B. (ο1994)，

Services in the Consumption Exper討ience，" in Rust, R. T. and Oliver R. L.,

(Eds扎 Service Quality: New Directions in Theory and Practice

,

(21-71)

,

Newbury Park

,

CA: Sage Publication.

Hsu, H. (2006),“An Empirical Study of Web Site Quality, Customer Value and

Customer Satisfaction Based on E-Shop," The Business Review, 5(1),

190-193.

Hsu, F. M., Chen, T. Y. and Wang, S. (2010),“The Role of Customer Values in

Accepting Infonnation Technologies in the Public Infonnation Service

Sector," The Ser叫celndustries Journal, 30(7), 1097-1111.

Hu, H. 旺， Kandampully, J. and Juwaheer, T. D. (2009),“Relationships and

Impacts of Service Quality, Perceived Value, Customer Satisfaction and

Image: An Empirical Study," The Service lndustries Journal, 29(2), 111-125.

Hume, M. and Sul1ivan Mort, G., (2010), “The Consequence of Appraisal

Emotion

,

Service Quality

,

Perceived Value and Customer Satisfaction on

Repurchase Intent in the Perfonning Arts," The Journal of Services Marketing,

24(2)

,

170-183.

Hung, W. T., Tsang, S. S. and Liu, H. Y. (2010),“Website Characteristics and the

Impavt of User Perceived Value on User Behavior in Web 2.0," Academy of

(27)

Lamb, R., King, J. L. and Kling, R. (2003), “Infonnational Environments:

Organizational Contexts ofOn1ine Infonnation Use

,"

Journal ofthe American

Society for Information Science and Technology

,

54(2)

,

97-114.

Jöreslog, K. D. and Sörbom, D. (1984), Analysis of Linear Structural Relationship

byMαximum Likelihood, Chicago, IL: Scientific Press.

Kettinger, W. J. and Lee, C. C. (1995),“Perceived Service Quality and User

Satisfaction with the Infonnation Service Function," Decision Sciences, 25(5),

737-765.

Kim, S. S. and Malhotra, N. K. (2005),“Predicting System Usage from Intention

and Past Use: Scale Issues in the Predictors," Decision Sciences, 36(1),

187-196.

Mason, R. O. (1978), “Measuring Infonnation Output: A Communication

Systems Approach," Information & Manageme刑， 1(5),219-234.

Nysve凹，旺， Pedersen, P. E. and Thorbjomsen, H. (2005),“Intentions to U se

Mobile Services: Antecedents and Cross-Service Comparisons," Journal of

the Academy of Marketing Scie前e，33(3), 330-346.

Oliver, R. (1997), Sati，φction: A Behavioral Perspective on the Consumer, New

York

,

NY: McGraw-Hi11.

Pihlström M. and Brush, G. J. (ο20∞08)，

Infonnation and Entertainment Mobi1e 臼ervl此ce試S s丸，" Psychology

&

Marketing,

25(8)

,

732-755.

Pitt, L. F., Watson, R. T. and Kavan, C. B. (1995),“Service Quality: A Measure

of Infonnation Systems Effectiveness," MIS Quarter，鈔， 19(2), 173-188.

Pope院， N. (1998),

(28)

102

E一一

Exp/oring Perceived Va/ue and Usage

01

Pura, M. (2005),“Linking Perceived Value and Loyalty in Location-based Mobile

Services," Managing Service Quali紗，的 (6)， 509-538.

Robbins, S. P. (2003), Organizational Behavior, 10th Edition, Englewood Cliffs,

NJ :Prentice-Hall.

Seiders, K., Voss, G. B. Grew祉， D. and Godfrey, A. L. (2005),“Do Satisfied

Customers Buy More? Examining Moderating Inf1uences in a Retailing

Context," Journal of Marketing, 69(4), 26-43.

Shannon, C. E. and Weaver, W. (1949), The Mathematical Theory of

Communication, Urbana, IL: University of Illinois Press.

Sheth, J. N., Newman, B. 1. and Gross, B. L. (1991),“Why We Buy What We

Buy: A Theory of Consumption Values," Journal of Business Research, 22(2),

159-170.

Sh此la， P. (2010),“Effects ofPerceived Sacrifice, Quality, Value and Satisfaction

on Behavioral Intentions in the Service Environment," Services Marketing

Quarter鈔， 31(4)， 466-484.

Stephens, D. O. (1998),“Megatrends in Records Management," ARMA Records

Management Quarter鈔， 32( 1)， 3-9.

Stewa肘， G. T. (2006), Re/ationa/ Va/ue Exchange: A Grounded Theory Study of

Buyers' and Sellers' Approaches to Building, Maintaining and/or Disso/ving Interorganizational Relationships

,

Unpublished Ph.D. Dissertation

,

The University ofTennessee.

Sullivan M。此， G. and Drennan, J. (2005),“Marketing M-services: Establishing a

Usage Benefit Typology Related to Mobile User Characteristics," Database

Marketing & Customer Strategy Management

, 12(4)

,

327-341.

Sweeney, J. C. and Sout缸， G. N. (ο2001) ，

(29)

Sweeney, J. 仁， Soutar, G. N. and Johnson, L.W. (1999),“The Ro1e of Perceived Risk in the Quality-va1ue Relationship: A study in a Retai1 Environment,"

Journal of Retailing, 75(1), 77-105.

Tha1er, R. (1985), “Menta1 Accounting and Consumer Choice," Marketing

Science, 4(3), 199-214.

Van Birgelen, M., De Jo月， A. and De Ruyter, K. (2006),“Multi-channe1 Service Retai1ing: The Effects of Channel Performance Satisfaction on Behaviora1 Intentions,\JournalofRetailing, 82(4), 367-377.

Venkatesh, V. and Davis, F. D. (2000), “A Theoretica1 Extension of the Techno1ogy Acceptance Mode1: Four Longitudina1 Fie1d Studies,"

Management Science, 46(2), 186-204.

Wang, J. S., Wu, 1. C. T. and Fu, H. 且. (ο2010)， H

Perceived Va1ue, Brand Reputation and Loya1ty: A Study on Traveling Industry," Journal of Informαtion & Optimization Sciences, 31(3), 659-678. Wang, Y. S. (2008),“Assessing E-commerce Systems Success: A Respecification

and Va1idation ofthe DeLone and McLean Mode1 ofIS Success," Information

砂stemsJournal, 18(5),529-557.

Wang, Y., Lo, H. P., Chi, R. and Yang, Y. (2004),“An Integrated Framework for Customer Va1ue and Customer-re1ationship-management Performance: a Customer-based Perspective 企om China," Managing Service Quali紗， 14(2), 169-182.

West, D. M. (2006), Global E-Government, 2006, Center for Public Po1icy, Brown University.

Wo∞od釘ruf託f， R. (1997),

(30)

104 Exp/oring Perceived Va/ue and Usage of Information Systems in Government Context

Zeithaml,

v.

(1988),

Means-End Model and Synthesis of Ev叫id由en郎1昀ce，"JournαalofM必σ抑r成keωtin咚1喀g， 52(3),