شخصیت بعنوان یک تعدیل کننده نظارت قبول واقعیت
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|38517||2003||15 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Computers in Human Behavior, Volume 19, Issue 4, July 2003, Pages 479–493
Abstract Organizational efforts at monitoring employee activity must be perceived as respecting privacy and fairness. However, even when monitoring systems are designed to do so, employees might not be willing to accept and use monitoring technologies. This study examined whether personality moderated the relationship between workplace monitoring system characteristics, fairness, privacy and acceptance. Six hundred and twenty-two participants were asked to provide their assessment of an awareness monitoring system (that determines employee availability to interact with geographically distributed colleagues) and to complete a five-factor measure of personality (i.e. extraversion, agreeableness, emotional stability, openness to experience, and conscientiousness). Results indicated that emotional stability and extraversion altered the relationships between the paths in a model of monitoring acceptance. Specifically, people who scored lower in extraversion and emotional stability were less likely to endorse positive attitudes toward monitoring, even with privacy and fairness safeguards in place. Implications for the expansion of models of workplace monitoring and for the practice of monitoring in organizations are discussed.
. Introduction Electronic monitoring of employee activities is becoming increasingly pervasive in organizations (Alge & Ballinger, 2001). Although numerous studies have examined the effects of electronic performance monitoring (EPM) on outcomes such as personal control (Stanton & Barnes-Farrell, 1996), productivity (Aiello and Kolb, 1995 and Larson and Callahan, 1990), privacy (Alge, 1999), fairness (Ambrose & Alder, 2000), and performance (Ambrose & Kulik, 1994), very few studies have looked at individual differences in predicting monitoring outcomes. Instead, researchers have investigated the influence of the situation (being monitored) and resulting outcomes (attitudes and behaviors). What is missing is information on individual characteristics. Thus, the question addressed in this investigation is whether individual difference variables—specifically personality variables—moderate the relationships between monitoring system characteristics (e.g. the frequency of monitoring) and outcomes such as perceptions of privacy, fairness and acceptance. Addressing this question will add an important piece of information to our understanding of reactions to monitoring. Although personality characteristics relate to important organizational outcomes (Funder, 2001), few studies have examined them in a monitoring context. One exception is a study by Douthitt and Aiello (2000): monitored individuals who were higher in negative affectivity reported lower levels of task satisfaction. Further, Robie and Ryan (1999) found that conscientiousness only predicted task performance when participants knew that they were being monitored. Finally, in a qualitative field study of video-based monitoring, Webster (1998) found that introverts were less likely to use these systems. Nevertheless, we still know very little about the moderating effects of individual difference variables on monitoring outcomes. Consequently, researchers such as McKnight and Webster, 2001 and Stanton, 2000 have called for further investigations into how personality variables might influence perceptions and attitudes toward monitoring. The present study seeks to examine these effects. 1.1. Personality and monitoring system acceptance This investigation concerning the moderating effects of personality focuses on reactions to awareness monitoring systems. These new monitoring technologies are being designed to enhance communications between geographically distributed colleagues (Lee & Girgensohn, 2002), rather than to measure job performance as with EPM. These awareness technologies operate on the principle that if the employee is aware of when his/her geographically distributed colleague is available to interact, the employee can be more effective in communicating with that colleague. Thus, a typical awareness monitoring system might capture a video-based image of the distributed colleague and transmit this image to the co-worker interested in engaging in communication (e.g. Erickson & Kellogg, 2000). Awareness systems have been implemented in organizations such as NYNEX and Xerox (Lee, Schlueter, & Girgensohn, 1997) and have been embraced by the computing community as an integral aid to collaborative work (e.g. Abowd and Mynatt, 2000 and Liechti and Sumi, 2002). Awareness continues to be designed into a variety of systems, including the Web, instant messaging, and wireless computing (Liechti and Sumi, 2002 and Weiss and Craiger, 2002). Despite this enthusiasm, researchers are only now beginning to recognize the difficulty in getting employees to accept and use these systems. There is a growing recognition that employees’ privacy concerns about being monitored for availability can have an impact on acceptance (e.g. Erickson et al., 2002 and Greenberg and Kuzuoka, 2000) and an acknowledgement that technologies that track presence and activity may lead to privacy concerns well beyond those elicited by performance monitoring (Weiss & Craiger, 2002). That is, despite the shift in purpose (i.e. from performance to non-performance tracking), the consequences of awareness monitoring for employees can be similar in nature to performance monitoring (Zweig & Webster, 2002). To examine the moderating role of personality, we drew on a base model developed by Zweig and Webster (2002) that examined whether system characteristics designed to enhance employee perceptions of privacy and fairness would result in more positive attitudes and greater intentions to use an awareness system (see Fig. 1). In their study of over 600 organizational employees, Zweig and Webster drew on research from EPM, fairness, privacy, and technology acceptance to justify the linkages among the constructs. For example, they drew on Leventhal's (1980) procedural justice rules to hypothesize that greater perceptions of fairness would result when: (1) images are captured and projected intermittently versus continuously, (2) images are blurred rather than clear, (3) the employee has control versus no control over who can access to awareness information and, (4) the employee is given knowledge of who is using the system to determine their availability versus no knowledge. Furthermore, they drew on theories around fairness (Ambrose & Alder, 2000), privacy (Eddy, Stone, & Stone-Romero, 1999), and technology acceptance (Davis et al., 1989 and Fishbein and Azjen, 1975) to hypothesize that greater perceptions of privacy and fairness would result in more willingness to accept the awareness technology, greater perceptions of its usefulness, and greater intentions to use the technology. Their analysis revealed that employees were very concerned about privacy and fairness when video-based images were used. However, modifying the characteristics of the awareness system to enhance perceptions of privacy and fairness (according to guiding principles outlined in EPM research) did little to mitigate negative reactions to the monitoring system and still resulted in low levels of acceptance. Here, we replicate and extend Zweig and Webster (2002) by collecting data on each of the variables in the model presented in Fig. 1, as well as on personality. Base model of awareness system acceptance (adapted from Zweig & Webster, 2002). Fig. 1. Base model of awareness system acceptance (adapted from Zweig & Webster, 2002). Figure options Personality refers to internal factors such as dispositions and interpersonal strategies that explain individual behavior, and the unique and relatively stable patterns of behaviors, thoughts, and emotions shown by individuals (Hogan, Hogan, & Roberts, 1996). Although there has been considerable debate about whether personality represents a stable characteristic that can usefully predict behavior or whether behavior is primarily determined by the situation, many researchers would agree that both individual and situational factors are important to understanding behavior (for a review, see Funder, 2001). Personality is often conceptualized in terms of a five-factor model, including the dimensions of extraversion, emotional stability, agreeableness, openness to experience, and conscientiousness (Funder, 2001, Goldberg, 1990, Goldberg, 1992 and McCrae and Costa, 1999). Thus, these five dimensions of personality were used to explore the moderating effects of personality in the base model. As mentioned earlier, few monitoring studies have investigated the influence of individual differences on attitudes toward monitoring. However, it has been recognized that individual difference variables such as personality might moderate the relationship between the characteristics of monitoring systems and attitudes toward monitoring (Stanton, 2000 and Webster, 1998). Next, we propose how personality might moderate the relationships diagrammed in Fig. 1. High levels of extraversion are reflected in traits such as sociability, gregariousness, and assertion (Barrick & Mount, 1991). Webster (1998) suggested that extraversion might relate to awareness system acceptance because those scoring lower in extraversion tend to have greater concerns for personal privacy. Thus, it is reasonable to expect that extroversion would moderate links with privacy invasion in Fig. 1, such that the links with privacy invasion will be stronger for those who score lower in extraversion. Although there is some specific evidence that extraversion might moderate some of the links in the base model, the other four elements of the five-factor model of personality also merit attention. For example, individuals with low levels of emotional stability tend to be defensive and guarded, have a negative view of themselves, worry about other's opinions of them, and tend to make stable, internal, global attributions about negative events (Clark & Watson, 1991). It is possible that for these people (as compared with those higher in emotional stability), being monitored would trigger concerns about how others view them resulting in stronger links between attitudes toward the technology and other variables. On the other hand, people who score higher on measures of agreeableness (e.g. non-competitive, cooperative and hopeful) might be more favorably disposed toward monitoring technologies that purport to enhance communication and cooperation with colleagues; thus, the links between attitudes toward the technology and other variables might differ as compared with their counterparts who score lower on agreeableness. As well, those who score higher on measures of openness to experience (e.g. imaginative, intellectual and curious) might be more willing to try out and use new technologies such as awareness systems; consequently, the links between attitudes toward the technology and other variables might differ as compared with those lower in openness to experience. The relationship between conscientiousness and awareness system acceptance is more complicated. The characteristics that describe high levels of conscientiousness—being hardworking, achievement oriented and persevering on tasks (Barrick & Mount, 1991)—suggest that those scoring high on this personality variable would welcome a tool that claims to enhance communication and performance. Thus, high levels of conscientiousness should be related to more positive attitudes toward monitoring. However, the desire for achievement that is characteristic of high levels of conscientiousness might result in concern over being constantly monitored by others, while those scoring lower on conscientiousness might not be concerned with having their actions monitored. To explore the relationships between personality and monitoring empirically, we examined whether differences in personality help to explain why some people respond negatively to monitoring technology, even when privacy and fairness are respected. To answer this question, and to address this lack of research, we chose to examine these relationships on an exploratory basis and offer the following research question: Research Question 1: Will personality moderate the relationship between the characteristics of a monitoring system, and perceptions of privacy invasion, fairness, attitudes and intentions to use the monitoring system?
نتیجه گیری انگلیسی
Results Table 1 presents the descriptive statistics, reliabilities, and correlations among the study variables. Table 2 presents the reliabilities for five personality variables across the entire sample as well as the sample size, means, and standard deviations for the same variables split into high and low groups. The model of awareness system acceptance presented in Fig. 2 exhibited a very high degree of fit with the data, (χ2 (19, N=664)=40.69, P=0.003; GFI=0.98, AGFI=0.96, RMSEA=0.04), and was consistent with results found in Zweig and Webster's (2002) employee sample. 2 Table 1. Descriptive statistics, reliabilities, and zero-order correlations among base model variables Variable Mean S.D. 1 2 3 4 5 6 7 8 9 1. Image Clarity 0.50 0.50 – 2. Frequency of Image Updating 0.43 0.50 −0.01 – 3. Knowledge of Monitoring 0.56 0.50 0.02 0.01 – 4. Control over Monitoring 0.57 0.50 0.04 0.02 −0.03 – 5. Fairness 3.63 1.31 0.01 0.02 0.12** 0.15*** (0.89) 6. Privacy Invasion 4.83 1.26 −0.02 −0.08* −0.08* −0.11** −0.62*** (0.84) 7. Attitude 3.10 1.38 0.05 0.08* 0.08* 0.05 0.75*** −0.60*** (0.91) 8. Intent to Use 0.33 0.42 0.02 0.02 0.02 0.09* 0.53*** −0.41*** 0.55*** (0.76) 9. Usefulness 3.97 1.34 0.03 −0.15** 0.02 −0.02 0.58*** −0.31*** 0.63*** 0.51*** (0.77) Numbers in parentheses are Cronbach's alpha estimates of internal consistency reliability. Means for independent variables (variables 1–4) represent the mean level for each independent variable (e.g. 1 for high, 0 for low). ∗ P<0.05. ∗∗ P<0.01. ∗∗∗ P<0.001 (two-tailed). Table options Table 2. Means, standard deviations and reliabilities for personality variables Variable Mean S.D. N Alpha (overall N=622) Agreeableness High 7.7 0.38 215 0.89 Low 5.8 0.57 204 Conscientiousness High 7.4 0.46 210 0.88 Low 5.1 0.57 213 Extraversion High 6.8 0.52 211 0.89 Low 4.3 0.65 219 Openness to Experience High 7.5 0.42 212 0.85 Low 5.4 0.54 205 Emotional Stability High 6.0 0.57 214 0.86 Low 3.7 0.54 206 Table options Results from the analysis of the base model. Note: +P<0.10, *P<0.05. Fig. 2. Results from the analysis of the base model. Note: +P<0.10, *P<0.05. Figure options Using this model as a baseline, we then examined the moderating effects of personality on each of the paths in the model.3 To do so, the SEM analysis was conducted twice for each of the Big Five personality variables—once with high levels of each personality variable and once again with low levels. This allowed for the calculation of overall fit statistics and the assessment of critical differences between the paths on the model for high and low levels of each personality variable (see Fig. 3). The fit statistics for the baseline and moderated models presented in Table 3 reveal that all of the models fit the data well. An examination of the critical ratios for differences among the paths for each of the personality characteristics is presented next. Moderator analyses on the base model—significant effects for Extraversion, ... Fig. 3. Moderator analyses on the base model—significant effects for Extraversion, Agreeableness and Emotional Stability. P<0.05 (one-tailed). *P<0.05 (two-tailed) for difference tests. Figure options Table 3. Fit statistics for baseline and moderated models Fit measure Baseline Extraversion Agreeableness Conscientiousness Openness Emotional Stability χ2 40.69 73.45 76.20 44.25 56.31 51.69 P 0.003 0.001 0.001 0.225 0.028 0.068 df 19 38 38 38 38 38 χ2/df 2.14 1.93 2.00 1.16 1.48 1.36 GFI 0.99 0.96 0.96 0.98 0.97 0.97 AGFI 0.97 0.92 0.92 0.95 0.93 0.94 RMSEA 0.04 0.05 0.05 0.02 0.03 0.03 Table options 3.1. Extraversion An examination of the differences among the paths for participants high and low in extraversion revealed that those scoring lower in extraversion expressed more negative relationships between several characteristics of the awareness monitoring system and perceptions of privacy invasion and fairness (see Fig. 3). Specifically, there were significant differences in the critical ratios for knowledge (of who is monitoring) and perceptions of privacy invasion (CR=−2.10; β=0.03, β=−0.19 for high and low extraversion, respectively). In other words, for those scoring higher in extraversion, knowledge of who is monitoring them does not matter. This is not the case for those lower in extraversion. For these people, knowledge of who is monitoring them was related to lower perceptions of privacy invasion. An opposite pattern was observed when examining control over the dissemination of monitoring information to colleagues and perceptions of fairness. Control was positively related to perceptions of fairness for those higher in extraversion but unrelated to perceptions of fairness for those lower in extraversion (CR=−2.13; β=0.11, β=−0.05 for high and low extraversion, respectively). While both those high and low on extraversion expressed a strong negative relationship between perceptions of privacy invasion and fairness, this relationship was significantly stronger for those lower in extraversion (CR=−3.38, β=−61, β=−0.86 for high and low extraversion, respectively). 3.2. Conscientiousness, openness to experience and agreeableness No significant differences were found among the paths in the models for conscientiousness and openness to experience. However, for agreeableness, two paths approached significance. Specifically, there was a weaker positive relationship between perceptions of fairness and attitudes toward monitoring for those lower in agreeableness (CR=−1.81; β=0.51, β=0.24 for high and low agreeableness, respectively). As well, attitudes toward monitoring were more strongly related to intentions to use the monitoring system for those scoring lower in agreeableness (CR=1.92; β=0.36, β=0.69 for high and low agreeableness, respectively) compared with those scoring higher (see Fig. 3). 3.3. Emotional stability For participants who rated themselves as higher in emotional stability, perceptions of the fairness of the monitoring system related more strongly to attitudes toward monitoring (CR=−2.91; β=0.56, β=0.12 for high and low emotional stability, respectively). In contrast, those scoring lower in emotional stability reported a stronger relationship between privacy invasion and attitudes toward monitoring than those scoring higher (CR=−1.91; β=−0.17, β=−0.40 for high and low emotional stability, respectively). As well, those lower in emotional stability expressed a stronger relationship between perceptions of the usefulness of the monitoring system and attitudes toward monitoring (CR=2.06; β=0.26, β=0.52 for high and low emotional stability, respectively) and between attitudes toward monitoring and intentions to use the monitoring system (CR=2.20; β=0.44, β=0.81 for high and low emotional stability, respectively). However, participants higher in emotional stability expressed a strong relationship between perceptions of usefulness and intentions to use the monitoring system (CR=−2.48; β=0.43, β=−0.03 for high and low emotional stability, respectively) while the relationship was not significant for those lower in emotional stability (see Fig. 3).