RUNNING HEAD: TECHNOLOGY USE, PRODUCTIVITY AND DISTRESS

Does Work-Related Technology Use Influence Employee Perceptions of Noelle Chesleya and Britta Johnsonb University of Wisconsin-Milwaukee Productivity and Distress?

Paper Prepared for Presentation at the First Inaugural Conference of the Work Family Research Network, NY, NY, June 14th 16th, 2012.

a Direct correspondence to: Noelle Chesley, Assistant Professor, University of WisconsinMilwaukee, Department of Sociology, 2025 E. Newport Avenue, Milwaukee, WI 53221; 414-229-2398 (phone); 414-229-4266 (fax); chesley@uwm.edu
b

Britta Johnson [4319 N. Alpine Ave, Shorewood, WI 53211; 414-517-4151 (phone); brittaeileen@aol.com]

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS Drawing on the Jobs Demands-Resources framework and 2008 survey data drawn Abstract

to higher levels of employee productivity and distress. We find that: 1) ICT-based work extension, telecommuting, and network expansion are linked to higher levels of underscore that how ICT is used is more important than whether it is used in understanding its impact on employees. Job-Demands-Resources model connected to higher productivity and lower distress levels. Overall, these findings

from a national sample of employees (PEW Networked Workers Study) , we test whether work practices driven by Information and Communication Technology (ICT) use are linked productivity and distress; and 2) Using ICT to complete a range of general work tasks is

Keywords: Information and Communication Technology (ICT) Use, Productivity, Distress,

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

organizations (Aneesh, 2006; Mamaghani, 2006). In spite of claims about the increased ICT use, we know relatively little about how work-related ICT use is influencing both

practices that are used to increase efficiencies and communication speeds, and that these technologies have created an infrastructure that supports the global expansion of

Technology (ICT) to perform their work (Bond, Thompson, Galinksy, & Prottas, 2002;

Madden & Jones, 2008). Research indicates that ICT has been incorporated into workplace

The majority of Americans use some type of Information and Communication

Does Work-Related Technology Use Influence Employee Perceptions of Productivity and Distress?

efficiencies of technology use, or concerns about unintended consequences associated with use support organizational goals, such as enhanced productivity? Are there any unintended consequences associated with work-related ICT use for employees, such as increased work practices to determine if these practices influence employee perceptions of productivity and distress. organizational and employee outcomes. How exactly is ICT used in workplaces? Does its

distress? The goal of this study is to address these questions by analyzing specific ICT-based There is evidence that the adoption of technological innovations in the work place

productivity levels and well-being. Previous research indicates that ICT broadens the 2003; Kurland & Bailey, 1999; Murray & Rostis, 2007; Valcour & Hunter, 2005). The

boundaries of where and when people can work (Bailey & Kurland, 2002; Bittman, Brown, & Wajcman, 2009; Duxbury, Towers, Higgins, & Thomas, 2006; Hill Ferris, & Martinson, blurring of work/non-work boundaries has been connected to increases in work-family

has changed the organization of work in ways that could influence both employee

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

conflict (Glavin & Schieman, 2012) and may increase employee productivity and/or distress (Chesley 2005). This should be especially true when ICT use occurs within organizational Rostis, 2007; Towers, Duxbury, Higgins, & Thomas, 2006). cultures that expect high levels of employee engagement and performance (Murray & Deepening our understanding of the role that ICT plays in shaping employee

outcomes requires more specific documentation of how employees use ICT at work and

Networked Workers Study (N = 1000)) from a nationally representative sample of workers. well as measures of employees perceived productivity and distress. This survey incorporates a range of measures tapping how ICT is used in workplaces, as Four sets of literature inform this project. We draw first on Jobs Demands-Resources

linking such practices to organizational and employee outcomes. To accomplish this, we

draw on survey data collected by the PEW Internet and American Life Project (The 2008

theory and incorporate findings from the interdisciplinary technology studies literature to Finally, we review research that links work-related ICT use to variations in employee productivity and distress. BACKGROUND better specify the potential role that ICT use may play in shaping work demands and resources. Next we describe what we know about how ICTs are used in the workplace.

and their interaction with each other is what determines employee health or distress. Job

proposes that job resources and demands are ongoing processes in the employment setting

characteristics of the work environment, like ICT use, to employee outcomes. This model

Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) provides one explanation linking

The Job Demands-Resources (JD-R) theoretical model (Bakker & Demerouti, 2007;

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS effects of job resources.

resources can buffer the impact of job demands, or job demands can override the positive resource for organizations and employees (Hill et al., 2003: Mamaghani, 2006; Valcour & The introduction of ICT into employment settings was intended to serve as a

Hunter, 2005), and some research supports this premise. For example, ICT-supported

telecommuting helps some workers manage work-life balance (Duxbury et al., 2006; Hill et al., 2003; Kurland & Bailey, 1999; Mamaghani, 2006), while ICT-based communication can improve the ability of employees to share their ideas with coworkers (Mano & Mensch, Accordingly, there is evidence that ICT-based work practices can serve, not just as unintended consequences and unanticipated (and often contradictory) effects (pg. 70). 2010). H owever, as Wajcman (2008) points out, technological innovations often generate

work-related ICT use to an increase in the pacing of work (Chesley 2010a; Green 2004a, 2010a; 2011) both of which are linked to increases in worker distress. Likewise, ICTOlson-Buchanan, 2007; Fenner & Renn, 2010).

2004b; Maume & Purcell, 2007), as well as a more fragmented work environment (Chesley facilitated work extension has been positively related to work-family conflict (Boswell & as a potential resource for employees, as well as a practice that may generate additional

resources, but may also promote greater work demands. Previous research has linked

demands. Thus, research that can better document the influence of ICT-based practices on employee outcomes can not only help us understand how to maximize the benefits and ICT-based practices operate as part of broad organizational systems.

The JD-R model theoretical model suggests that ICT use needs to be understood both

minimize the costs associated with these practices, but may also shed light on how specific

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS ICT-Based Work Practices Survey data from 2008 indicate that 62% of US employees use the Internet or email

(Desjardins Financial Security, 2006). While previous research recognizes the key role of ICT in organizations, past research often characterizes this use in very general terms, like frequency of computer or Internet use. Our focus is not on very general forms of ICT use, but on ICT use as it is embedded in a series of work practices that were not possible or could not be performed on the same scale before the entrenchment of computers and mobile communications (Chesley and Johnson, 2010b). Scholars have identified and

at their workplace (Madden & Jones, 2008), Similar patterns have been identified in Canada

studied some work practices that are intimately connected to ICT use, such as ICT-

facilitated work extension (supplemental work done at home in addition to working

traditional work hours; Duxbury et al., 2006; Fenner & Renn, 2004, 2010; Towers et al., 2006), ICT-supported teleworking (Bailey & Kurland, 2002; Hill et al, 2003; Kurland &

2008; Milliken & Dunn-Jensen, 2005), and ICT-based task completion (e.g. Email, instant Horvitz, 2000; Mano & Mensch, 2010, Mark, Gudith, & Klocke, 2008).

messaging, text messaging, etc.; Cutrell, Czerwinski, & Horvitz, 2001; Czerwinski, Cutrell, & The current literature lacks data on the incidence of some of the specific ICT-based

Bailey, 1999), ICT-facilitated social network expansion (Kennedy, Smith, Wells, & Wellman,

measured the amount of hours employees in their sample worked from home (i.e.

telecommuting) and performed unpaid supplemental work (i.e. work-extension). They find

findings difficult to apply to broad populations. A 2005 US Department of Labor analysis

work practices just described (i.e. ICT-based task completion measures). When ICT-based work practices are measured, inconsistent measurement strategies make using these

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

that about 15% of respondents worked from home regularly (20.7 million) and 10.2 million Americans reported that they performed job-related work at home without formal compensation. Of these two groups, roughly 80% of these workers reported that they had access to ICT devices that facilitated their work at home (US Department of Labor, 2005). Job-type is associated with the level of work-related ICT use with almost three-fourths of Limited previous research also provides some guidance on frequency of ICT use.

professionals, executives, or managers reporting use of the internet at work constantly or several times a day (Madden & Jones, 2008). F urthermore, white-collar employees own

and use more gadgets such as laptops, PDAs, and mobile phones than other employees, and Overall, more information is needed about how employees use ICT at work, not what ICTs they use, or how frequently they use specific devices. Previous research does suggest a relationship among ICT adoption and broad

tend to use them more frequently both in and out of the workplace (Madden & Jones, 2008).

ICT-Based Work Practices and Employee Productivity

workplace productivity gains (Aral, Brynjolfsson, & Wu, 2006; Baily, 2004; Brynjolfsson & relatively little research to date has looked at how ICT use impacts individual levels of analyze, and distribute information (ODriscoll, Biron, Cooper, 2009) which may help employee productivity. ICT use provides employees with new ways to capture, organize,

Hitt, 2000; Stiroh, 2002) as well as organizational efficiency (Mamaghani,2006). However,

In a 2001 Canadian survey, respondents linked ICT use to greater work productivity and

interest in their work (Duxbury & Higgins, 2001). Further, drawing on two regional samples of U.S. employees, Chesley (2010a) found that frequency of ICT use (computer, email, and

increase individual productivity by creating more efficient means to complete work tasks.

and Mensch (2010) use a nationally representative sample of U.S. employed adults to Hill et al. (2003), found that telecommuters have higher levels of self-reported job productivity than their colleagues in other work arrangements. Another way work-related ICT use can lead to productivity increases is by

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

cell phone) is positively associated with levels of perceived workplace effectiveness. Mano analyze the effects of email on employee perceptions of work effectiveness and found that the amount of email sent and received is positively related to work performance. Finally,

facilitating practices that extend the workday. There is a growing body of research that has data (BlackBerry and Intel) reporting that users of notebook computers at home work an additional 5 hours per week, which results in an annual dollar benefit of $19,200 per Canadian government department about their ICT use after hours and found a positive studied this phenomenon (e.g. Boswell & Olson-Buchanan, 2007; Duxbury et al., 2006;

Fenner & Renn, 2010, Towers et al., 2006). For example, Duxbury et al. (2006) cite industry employee for the organization (p. 11). Towers et al. (2006) surveyed employees in a major

relationship between frequency of ICT use outside of the office and total hours worked each week. subjective measures of productivitysome argue that subjective measures of Up for further debate within this literature is the argument about objective versus

productivity/performance are not an accurate way to assess this research concept. However, subjective measures of performance are used in research and are often 2004). On a firm-level, there is evidence that subjective and objective measures of

interpreted as being equivalent to objective measures (Wall, Michie, Patterson, & Wood, performance are positively associated (Wall et al., 2004). Overall, previous research has

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

documented a positive association between work-related ICT use and higher productivity ICT-based work practices result in different productivity outcomes. Does ICT Use used in specific ways to address particular types of work tasks? ICT-Based Work Practices and Employee Distress using both subjective (e.g. self-perceptions) and objective (e.g. actual hours worked)

measures. One area that needs further attention is investigating whether different types of encourage productivity across the board, or do productivity gains require that ICTs are

& Higgins, 2001; Mano & Mesch, 2010; Murray & Rostis, 2007). In a 2006 survey of Security, 2006). Similarly, Duxbury et al. (2006) reported that 45% of Canadian

Canadian workers, 54% of workers reported that ICT use maintained their existing level of respondents believed that ICT use had increased their stress levels. An under-studied area co-workers, for example) might underlie specific employee outcomes, such as distress. A limited number of studies have attempted to understand this in more detail. For

Work-related ICT use has been linked to employee distress (Chesley, 2005; Duxbury

stress, while 29% reported that ICT use increased their level of stress (Desjardins Financial of scholarship is how specific forms of ICT use (using email to communicate with clients and

example, Murray & Rostis (2007) suggest that it is the constant communication delivered Fernandez-Montejo (2010), studied the effects that communication-oriented tasks associated with stress (systolic blood pressure and heart rate). They found that

by ICT and the inability to disconnect that is stressful for ICT users. Ferrer, Rosa, Abad, and

completed via computer versus in person had on cardiovascular response markers typically versus face-to-face, suggesting a link between ICT use and physiological stress markers for

respondents had greater cardiovascular responses when they completed tasks via computer

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS some ICT-related tasks.

spaces and times which leads to negative effects on psychological well-being such as stress and distress (Chesley, 2005, Duxbury et al., 2006; Major, Klein, & Erhart, 2002). Another been shown to increase work hours for some ICT users (Boswell & Olson-Buchanan, 2007;

Some scholars argue that it is the ICT-facilitated intrusion of work into non-work

10

levels of perceived stress (Major et al., 2002) and clearly links higher stress levels to poor physical and mental health (Thoits, 2010). related ICT use and employee perceptions of productivity and distress. There is limited evidence linking specific ICT devices/platforms and frequency of use to both employee 2010). Other research focuses on use of a range of devices or applications on user

Duxbury et al., 2006; Fenner & Renn, 2010, Towers et al., 2006). This may be detrimental

potential link between ICT use and levels of distress/stress in employees is how ICT use has

for workers as previous research indicates that working longer hours can lead to increased Overall, previous theory and research clearly support a connection between work-

productivity or distress (e.g. Bittman et al., 2009; Czerwinski et al., 2000; Mano & Mesch, perceptions and work outcomes (e.g. Boswell & Olson-Buchanan, 2007; Chesley, 2010a; of workplace practices, like using ICT to manage contacts, schedule meetings, or work Murray & Rostis, 2007). However, we are not aware of research that examines how a series

outside the office, influence measures of employee productivity and distress. Using the JD-R framework as a guide, the current project will further clarify the relationship between ICTinformed by previous research and testing them will allow us to confirm and further based work practices and individual employee outcomes. The following hypotheses are

specify the relationships, if any, among ICT-based work practices and employee

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS productivity and distress: H1: Employees that are engaged in ICT-based work practices will report higher levels of productivity compared to those who do not engage in these practices, controlling for other factors. levels of distress compared to those who do not engage in these practices, controlling for other factors. METHOD H2: Employees that are engaged in ICT-based work practices will report higher

11

Data

the PEW Internet and American Life Project. The focus of the survey was to determine the in influencing levels of productivity and distress in employed US adults because of the

amount, type, and influence of technology use on a nationally representative sample of US breadth of ICT related topics it covers and because the data are drawn from a recent and computer, internet, email, instant messaging, or cell phone use. Our analytic sample

This study uses survey data from the 2008 Networked Workers survey conducted by

workers. This is an ideal dataset for examining the role that ICT-based work practices play nationally representative sample of US working adults. All individuals interviewed for this includes all respondents who indicated that they use some form of ICT and excludes cases

survey (n = 1000) self-identified as full-time or part-time workers. ICT use was defined as with missing data from the analysis, resulting in an analytic sample of 712 workers.

not a consistent relationship between response rates and non-response bias. A recent

The response rate for this survey was 24%. Recent research suggests that there is

with measures of key auxiliary variables (particularly gender and urban/rural address) on used auxiliary variables (including gender and region) to create a weight to correct for

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

respondents and non-respondents are best positioned to deal with issues of response bias.

meta-analysis (Groves 2006) suggests that data drawn from probability-based samples

12

Our data are drawn from a probability-based sample with a clear sampling frame and PEW response bias and Taylors linearization method is used to incorporate a design effect into

differential patterns of non-response. In our analyses we use this weight to correct for nonin STATA 11.2 (e.g. svy: logistic) were used to produce our estimates. Measures the standard error estimates (following recommendations from PEW). Survey commands

email, cell phones, instant messaging improved your ability to do your job? The responses range from 1 (a lot) to 4 (not at all). This variable is recoded as a dichotomous variable where 1 indicates improvement in productivity (some, or a lot) and 0 indicates little to which asks How much, if at all, have technologies such as the internet, email, cell phones, instant messaging increased stress in your job? The responses ranged from 1 (a lot) to 4 increase in distress (some, or a lot) and 0 indicates little or no increase in distress. no increase in productivity. Perceived employee distress is measured through a single item

question that asks respondents, How much, if at all, have technologies such as the internet,

Dependent Variables. Perceived employee productivity is measured through a single

(not at all). This variable is recoded as a dichotomous variable where 1 indicates an

is measured using two index variables that tap work extension due to email or phone use.

Key Independent Variables: ICT-Based Work Practices. ICT-facilitated Work Extension

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

Respondents were asked: Do you check your work-related email on: 1) weekends; 2) Responses range from 1 (often), to 4 (never). Responses were recoded so that higher

vacations; 3) before work; 4) after work; 5) when you are sick; 6) when running errands? numbers indicated greater work extension and summed together to form an index (6 = no email-related work extension; 24 = high email work extension). Respondents were also asked an identical set of questions about phone use. A general index of work extension 48 = high work extension). combines all measures of email and phone use outside of work (12 = low work extension; Has using email changed the amount of time you spend: 1) specifically working from Measures of ICT-Based Telecommuting are based on responses to two questions:

13

home, and 2) working at places other than the office and at home? Responses for both not use work email are grouped in the third comparison category (no change).

questions are: 1) yes, increased; 2) yes, decreased; 3) no, has not changed the amount of

time I spend doing these things. These variables were recoded so that respondents who do ICT-Based Network Expansion is measured using responses to a single item: How

much, if at all, have technologies such as the Internet, email, cell phones, instant messaging correspond to greater levels of network expansion (1= no network expansion to 4 = a lot of series of specific workplace tasksglobally referred to as ICT-Based Task Completion. Respondents were asked: Which is the most effective way to handle the following network expansion). A last set of variables captures the role of ICT use in effectively accomplishing a expanded the number of people you communicate with? Responses are: 1) not at all; 2) only a little; 3) some; 4) a lot. This variable was recoded so that higher numbers

about work issues; 4) deal with sensitive issues; 5) bring a problem to your supervisor? From this, we created a series of five binary variables that capture whether a task (e.g., perceptions of productivity and distress. Gender, age, race, and level of education are arranging meetings) is best accomplished in person (0) versus using ICT (1).

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

situations: 1) arrange meetings/appointments; 2) edit/review documents; 3) ask questions

14

into four racial/ethnic categories (white, black, Hispanic, other), small cell sizes require us to collapse race into a dichotomy (0 = other, 1 = non-Hispanic white). Education is also than high school; 2) high school graduate; 3) some college; 4) college or better. controlled and measured categorically; respondents fall into one of four categories: 1) less employment that may have an impact on the dependent variables. To measure hours In addition, we control for several characteristics associated with a respondents

factors known to influence job conditions at work. We control for gender (1 = female), age (measured in years), and race/ethnicity. Although respondents were originally classified

Control Variables. A number of other characteristics are likely to influence employee

worked, respondents were asked, How many hours do you work in a typical week? The responses ranged from 1 to 61 (PEW coded everyone working more than sixty hours per

wage last year (2007) before taxes? The original variable had eight response categories, The income variable had a fairly large number of missing cases (103 or about 10%). 1) Less than $30,000; 2) $31,000 to $49,000; 3) $50,000 to $74,000; 4) $75,000 or more.

week as 61, thus this variable is truncated). The income measure asks, What was your

several with small cell sizes. As a result, we recoded this item into a four-category variable:

Multivariate analyses of the patterns of missing income data indicate that age (p < .001) and education level (p < .05) are positively related to missing income reports (analyses not

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS shown).

productivity and distress experiences. We control for job type using a categorical measure based on the question, What type of work do you do? with possible responses: 1) Professional; 2) Manager/Executive/Business Owner; 3) Clerical/Office/Sales; 4) Service business owner into the Manager/Executive category and semi-skilled trades, and binary measure (1=self-employed; 0=otherwise). Finally, we control for type of other into the skilled trades category. We also control for self-employment using a Work; 5) Skilled Trades/Semi-Skilled Trades/Other. Due to small cell sizes, we combined

Other job and organizational characteristics are also known to influence

15

medium-size company; 3) small business; 4) federal/state/local government; 5) school or educational institution; 6) non-profit; 7) other. This variable was reduced to four government, school/educational institution, non-profit, other). sized company); 3) Small Private Sector (small business); 4) Other (federal/state/local The remaining variables draw on job characteristics that measure employee

organization/employer. There are seven response options: 1) large corporation; 2)

categories: 1) Large Private Sector (large corporation); 2) Medium Private Sector (medium-

perceptions of job satisfaction, autonomy, advancement opportunities, and complexity. To measure job satisfaction, respondents were asked, How satisfied are you with your job? recoded as a binary variable to reflect those that are generally satisfied (1) and those that are

The responses ranged from 1 (completely satisfied) to 4 (completely dissatisfied). This was generally dissatisfied (0). Job Autonomy is measured with a single item which asks, I have a lot to say about what happens in my job. Responses range from 1 (strongly disagree) to 5

(strongly agree). Job Advancement is also measured using a single item. Respondents were

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

asked I have opportunities for advancement in my job. Responses range from 1 (strongly disagree) to 5 (strongly agree). Response scales for both Job Autonomy and Job Advancement were reversed so higher numbers correspond to higher levels of agreement. skill 2) My job requires creativity 3) My job requires that I do the same thing over and over and 4) My job requires abstract knowledge about the ideas behind my work The (except for the question that asks, My job requires me to do the same thing over and over). The final measure is an index that sums across these four items (4 = low job complexity to 20 = high job complexity). responses for these questions range from 1 (strongly disagree) to 5 (strongly agree). The variables were recoded so that higher numbers correspond to higher levels of agreement

16

Job Complexity is measured using responses to four items: 1) My job requires a high level of

Analytic Strategy

that test the influence of each ICT-based work practice on productivity or distress after accounting for all control variables. These results are summarized in Tables 2 & 3. RESULTS

of each ICT-based work practice on the dependent variables, we estimate a series of models

work practices, perceived productivity, and distress. In order to fully capture the influence

We use bivariate logistic regression to analyze the relationships among ICT-based

workers report an increase in distress that they believe is connected to ICT Use, while more than two-thirds report that their productivity has increased as a result of their ICT Use. Thus, a sense that ICT use is connected to productivity is more prevalent than the sense

Table 1 provides descriptive statistics for the analytic sample. About a third of these

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

workforce. The average work extension score is about 22, which suggests that using email and phones to access work outside of standard times and places (e.g. while sick or on vacation) is ubiquitous. Indeed, only 13% of our sample reports never using the phone or email to check in with work on weekends, before or after work, etc. Most contemporary workers appear to use technology to communicate with work in places that used to be outside of the reach of work (like running errands). [Table 1 About Here]

that use is stressful in this national sample of workers. We can also get a sense of how prevalent particular ICT-Based work practices are in a recent cross-section of the

17

they spend telecommuting. Less than a fifth of the workforce (18%) reports an increase in the time they spend using technology to facilitate working at home or someplace else. norm. Respondents do agree that ICT use is influencing the size of their work-related scope of their work related communications. Thus, increases in ICT-based telecommuting (from home or other locations) are not the

The majority of respondents report that ICT use has not changed the amount of time

networks. When asked to rank their network expansion on a scale of 1 (not at all) to 4 (a Finally, the measures for ICT-facilitated task completion indicate a strong

lot), the mean response is 3, which suggests that most workers have seen an increase in the preference to use ICT to complete specific work tasks. For example, 52% of respondents 14% who prefer to do these things in person and 34% who do not use ICT at work), and

use ICT (email, phone, IM, or text message) to arrange meetings or appointments (versus

46% use ICT to edit or review documents (versus 21% in person and 33% who do not use

ICT) . However, some work tasks are still primarily attended to in person, including: asking

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS problem to your supervisor (54%).

questions about work issues (45%), dealing with sensitive issues (59%), and bringing up a average employee age is 40. The majority of the sample is white (77%) and fairly well Demographically, just under half of the respondents are female (47%), and the

18

educated; 41% of respondents have a college degree or better. The average number of

hours worked per week is 41 and the largest income grouping is those that make less than and professional (22%). The majority of the respondents are employed by someone else (88%) versus 12% who report being self-employed. The sample largely works in the private sector (74%) compared to 26% that work in other sectors. (27%). The two most common job type categories were skilled/semi-skilled/other (31%) $30,000 per year (35%), followed closely by those who make $30,000 to under $50,000

this sample. For example, 90% of respondents reported they were completely or mostly satisfied with their job. The mean for job autonomy is 3.7 (on a five point scale) and the and skill. is 14, which suggest many respondents are in jobs that require greater levels of creativity ICT-Based Work Practices, Distress, and Productivity The influence of ICT-based work practices on distress and productivity are mean for job advancement opportunities is 3.4 (on a 5 point scale). Average Job complexity

Job satisfaction, job autonomy, and job complexity measures were all fairly high for

documented in Table 2. The first set of models tests the influence of phone and email based work-extension on employee self-reports of increased distress and productivity. The results illustrate that both phone and email use can extend work outside traditional work boundaries in ways that are associated with increased distress and productivity. For each

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

one-unit increase in work-extension, perceived distress increases about 5% (p < 0.001) and perceived productivity increases about 9% (p < 0.001). [Table 2 About Here] We also tested the influence of telecommuting on employees reported distress and

19

productivity levels. After controlling for a number of demographic and work-related factors, the models suggest that increased ICT-facilitated commuting (from home or another location) is positively linked to perceived distress and productivity levels. Here, employees experiencing an increase in time spent telecommuting are 124% more likely to report an

increase in distress when compared with employees who report no change in telecommuting time. Increases in telecommuting are also associated with perceived productivity increases. Here employees reporting a telecommuting increase are also 589% more likely to report an in both distress and productivity. Employees experiencing a one-unit increase in network expansion are 43% more likely to report increased distress related to ICT use (p < 0.001) and 119% more likely to report productivity gains related to work-based ICT use (p < .001). Our final set of models on table 2 tests the influence of doing more general workincrease in productivity. ICT-facilitated network expansion is also associated with increases

based tasks (e.g. arranging meetings, reviewing documents, etc.) in person rather than using technology on employee reports of increases in distress or productivity. These models indicate that doing more of these tasks in person increases the likelihood that an employee also reports increased distress by 20% (p < 0.01). There is no evidence that performing more of these tasks using technology is linked to increased employee distress. The second model in this set illustrates that doing more of these tasks generally predicts self-assessed

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS to productivity gains.

productivity increases. However, the Odds Ratios for both the in-person and technology To further explore the role of completing these tasks in person versus using

variables indicate no practical difference in how these tasks are completed when it comes

20

technology, we estimated a series of models that assess the separate influence of each of

these tasks on perceived employee distress and productivity. Table 3 reports just the Odds Ratios, standard errors, and significance of the ICT work-task variables from models that

included all of the control variables documented in Table 2. The general pattern in this table illustrates that completing all of these tasks at work predicts both higher distress and higher productivity regardless of whether the tasks are completed in person or using technology.

However, the effect sizes tend to be larger for productivity than distress. In other words, an employee is more likely to experience an increase in productivity than distress when completing these tasks. [Table 3 About Here]

likely to lead to increased employee distress than completing these tasks using ICT. Indeed only in the case of bringing a problem to a supervisor is the Odds Ratio for in-person when considering task completion and productivity. Here, only two tasks (arranging productivity increases if they are accomplished via ICT rather than in-person. completion smaller than that for ICT completion. This sort of pattern is not as apparent

We can also see that, on balance, completing many of these tasks in person is more

meetings/appointments and bringing a problem to a supervisor) are more likely to lead to

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

of general Internet or computer use to better understand the context that underlies workrelated ICT use in todays workplaces. Our analysis of the 2008 PEW Networked Workers outside of traditional work boundaries (e.g., work extension) is common, but increased data tells us more about the range of workplace practices ICT use supports, drawing on a time spent working outside the office (e.g., ICT-based telecommuting) is not. In addition, cross-section of employees. We find that some level of work-related email and phone use

One contribution of this paper is to move beyond global measures of the frequency

DISCUSSION AND CONCLUSION

21

we find that, on average, employees report expanded work-related networks that are supported by ICT use. The role of ICT use in facilitating network expansion is well-

documented (e.g., Kennedy et al., 2008), so this is not surprising. Perhaps more informative are the patterns that document whether employees complete a range of common work tasks in person or using ICT. Here we find that using ICT to facilitate coordination and issues or problems in person, rather than via communications technology. Taken as a communication in some key work contexts. whole, we believe these descriptive patterns describe work environments in which editing tasks is common, but that many employees still ask questions, or raise sensitive technology is augmenting more routinized tasks, but is not replacing the role of in-person A second contribution of this study is documentation of associations among a

range of ICT-based work practices and two key employee outcomes: distress and

productivity. Do ICT-based work practices, like those we describe above, support

organizational goals such as enhanced productivity? The general answer is yes. Across our

statistical models, the overall pattern that emerges is that a range of ICT-based work practices

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

are positively associated with employee perceptions of greater productivity. This is consistent with the notion that ICT use is a key organizational resource that provides employees with the tools they need to enhance their work performance. Extending the reach of work via ICT perceptions of increased productivity.

22

through enhanced, but informal connectivity (like checking email on vacation or telecommuting) and the creation and maintenance of bigger work-related networks is associated with employee On the flip side, however, we do see evidence of unintended consequences

in distress levels. This is especially illustrated by the findings for ICT-facilitated work is linked to self-reported increases in both productivity and distress for employees,

extension, telecommuting, and network expansion. Each of these ICT-based work practices other words, while checking email on vacation may be great for organizations, it may not be so great for employees. Further, the increased connections facilitated by ICT use can provide individuals with wider communication networks which may be helpful when

associated with work-related ICT use for employees that may come in the form of increases

suggesting that ICT-based productivity may come at the expense of employee well-being. In

information is sought, but could prove to be stressful when an individual is inundated with information or requests. Much of the research examining the implications of ICT use for network expansion has focused on personal ICT use and has largely found that ICT use

facilitates social connection in ways that are positive for users. However, our findings potential cost to the employee in the form of increased distress.

illustrate the double-edged sword nature of ICT use for work-related network expansion by documenting that work-related use has the capacity to facilitate productivity but at a There are some interesting conclusions that follow from linkages among general task

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

completion that incorporates ICT use and increases in employee distress and productivity.

in ways that ICT facilitated task completion is not. Further, whether one generally completes a increases.

Models that examine the overall influence of generally completing tasks in person vs. using ICT

23

range of tasks in person or using ICT makes little difference in terms of perceived productivity

suggest that in-person task completion tends to be associated with increased employee distress

some of these work tasks, ICT use appears to lead to higher levels of perceived productivity that increasing ICT use for completion of specific work-related tasks may serve as a while also being linked to lower odds of being distressed. These overall patterns suggest resource for employees in ways that enhance productivity and minimize distress. Of course, it is also possible that some communications tasks (e.g. attending to sensitive issues, problems, account for in our models (problem/question/issue difficulty). Further exploring these true, then the patterns we report may be an artifact of something we do not measure and and questions) are more likely to be completed in person, rather than using technology. If this is

when asking questions of someone further up the organizational hierarchy or dealing with

distress. Using technology to mediate communications with supervisors or coworkers may allow employees to create workplace barriers that are protective in terms of well-being, particularly

On the one hand, it may be that ICT use has protective effects when it comes to employee

sensitive issues, but that have no consequences in terms of employee productivity. Thus, for

theoretical framework in order to better understand the role of ICT use in promoting

associations is a direction for future research.

particular organizational goals. A key take-away point from this study is that some ICT-

As a group, our findings also point to the need for further refinement of the JD-R

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

based practices (work extension, telecommuting, network expansion) appear to

simultaneously serve as both an organizational resource and a source of greater demands for employees. The point is that work-related ICT use is complex in its influence on work processes and outcomes and use is critically linked to a range of work practices. As such, ICT use will not be easily classified as either a general resource or demand across employees, industries, or jobs, which makes it difficult to incorporate into the JD-R

24

framework as a single variable. It may be that a focus on more specific ICT facilitated work practices that are used across industries or settings will be more useful for thinking about resources.

the many ways (both positive and negative) ICT use influences organizational demands and sectional research design used in the PEW data collection establishes association but is not able to determine causality. While our analysis assumes that employing specific ICT-based Our findings should be interpreted with several limitations in mind. First, the cross-

work practices influences subsequent levels of employee productivity or distress, it may be of ICT in workplaces likely dictates that the majority of workers have to use some type of

that more productive or distressed employees are more likely to use ICT at work. However, two pieces of evidence dispute the latter direction of the findings. First, the widespread use

ICT in order to complete work tasks--thus ICT use is probably no longer a choice, but a job the dependent variables (not every ICT based work practice is related to productivity or distress), which suggests that nuances in work-related ICT use do influence employee outcomes, even among ICT users.

requirement in the vast majority of settings. Second, there is variation in the findings across

A second limitation involves missing income data. Approximately 100 respondents

income; analysis not shown). This analysis indicates that older and less educated

coefficients in regression analyses. To assess the potential for bias, we regressed all of the

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

lower levels of education are generally associated with less frequent technology use and less technological skill, but it is difficult to know how these patterns might bias our findings. One possibility is that the loss of these cases from our analysis makes our distress) without being able to best capitalize on their benefits (e.g. productivity). estimates here more conservative, since employees with less technological skill are

respondents were more likely to refuse to give an income estimate. Both older age and

control variables on a binary measure of missing income (1 = income missing; 0 = reported

did not report their income when completing the PEW survey, a shortcoming that can bias

25

probably more likely to suffer the ill effects associated with work-related ICT use (e.g.

common concern when response rates are low is that generalizability of the sample to the population is in doubt. However, the most recent research indicates that low response rates alone are not a solid indicator of data quality (Groves 2006). A more worrisome source of bias is social processes that influence both participation in a survey and the

A third potential limitation of this study is the low survey response rate (24%). A

outcome being measured. In this study, it is possible that individuals with high distress understate the size and scope of an ICT use distress relationship.

levels would be less likely to participate in the PEW survey. However, if such a selection

process were in place it suggests that our results for distress are conservative and perhaps it is important to produce estimates based on nationally representative samples of workers, As a whole, these findings clarify the direction of future research in this area. While

as we do here, future research focusing on specific industries or professions could better

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

among high performance work systems (HPWS), ICT-based work practices, and employee initiatives that seek to engender high levels of employee effort, commitment, involvement, and identification with company objectives through specific work and employment Kalleberg, Marsden, Reynolds, & Knoke, 2006; Macky & Boxall, 2008). productivity and distress. HPWSs, also referred to as high commitment (Walton, 1985) or, high involvement (Lawler, 1986) work systems, are human resources management

specify whether and how specific ICT-based practices influence a range of employee

outcomes. An additional area that warrants more research is further examination of a link

26

practices (Appelbaum, Bailey, Berg, & Kalleberg, 2000; Boxall & Macky, 2009; Green, 2000; suggest that ICT use can help to enable that engagement. We show that ICT does enable

expanded connections (via network expansion). It is also used to facilitate completion of a host of specific work tasks in ways that appear linked to greater productivity. Thus, our

greater access to employees outside of traditional boundaries (work extension), and

These systems seek to heavily engage employees in work processes and our findings

findings highlight patterns that have specific implications for better understanding the role that technology use may play in shaping the culture, organization, and implementation of exploring this linkage here. Even with its limitations, this study provides new information about how specific High Performance Work Systems (HPWS), although data limitations keep us from further

ICT-based work practices, rather than overall frequency of ICT use, influences two key

how ICT is being used in the workplace is critical if we want to better understand the role of

such an array of ICT-based work practices in one study. Research that can better specify

employee outcomes: productivity and distress. We know of no other research that analyzes

provides researchers and organizations with better information about the social and organizational implications of work-based ICT practices.

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

technological innovation in organizations. This study helps to fill this knowledge gap and

27

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

REFERENCES

Applebaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. (2000). Manufacturing advantage: Why Aral, S., Brynjolfsson, E., & Wu, D. J. (2006, December). Which came first, IT or productivity? International Conference of Information Science (ICIS), Milwaukee, WI. The virtuous cycle of investment and use in enterprise systems. Paper presented at the high-performance work systems pay off. Ithaca, NY: ILR Press.

Aneesh, A. (2006). Virtual Migration. Durham and London: Duke University Press.

28

Bailey, D. E., & Kurland, N. B. (2002). A review of telework research: Findings, new Behavior, 23, 383-400.

Baily, M. N. (2004). Recent productivity growth: the role of information technology and other innovations. Federal Reserve Bank of San Francisco Economic Review. economics/review/2004/er35-42bk.pdf Retrieved September 2011, from http://www.frbsf.org/publications/ Journal of Managerial Psychology, 22, 309-328.

directions, and lessons for the study of modern work. Journal of Organizational

Bakker, A. B. & Demerouti, E. (2007). The job demands-resources model: State of the art. time pressure. Work, Employment, and Society, 23, 673-691.

Bittman, M., Brown, J. E. & Wajcman, J. (2009). The mobile phone, perpetual contact, and of the Changing Workforce. New York, NY: Families and Work Institute.

Bond, J. T., Thompson, C., Galinsky, E., & Prottas D. (2002). Highlights of the National Study Boswell, W. R., & Olson-Buchanan, J. B. (2007). The Use of Communication technologies Management, 33, 592-610. after hours: The role of work attitudes and work-life conflict. Journal of

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS 19, 3-23.

Boxall, P. & Macky K. (2009). Research and theory on high-performance work systems: Brynjolfsson, E., & Hitt, L. M. (2000). Beyond computation: Information technology, Perspectives, 14(4), 23-48.

Progressing the high-involvement stream. Human Resource Management Journal, organizational transformation, and business performance. Journal of Economic

29

Chesley, N. (2005). Blurring boundaries? Linking technology use, spillover, individual and pace of life. Information, Communication & Society, 13, 485-514.

Chesley, N. and B. Johnson. (2010b). Information and Communication Technology, Work, and Family. Pp. 1 36 in S. Sweet and J. Casey (Eds.) Work and family encyclopedia. Chestnut Hill, MA; Sloan Work and Family Research Network.

Chesley, N. (2010a). Technology use and employee assessments of productivity, workload,

distress and family satisfaction. Journal of Marriage and Family, 67, 1237-1248.

Chesley, N. (2011). Technologically tethered workers? Technology use, job conditions, Cutrell, Edward, Mary Czerwinski, and Eric Horvitz. 2001. Notification, disruption, and Czerwinski, Mary, Edward Cutrell, and Eric Horvitz. 2000. Instant messaging and Australia. presented at the IFIP Conference on Human-Computer Interaction, Tokyo. Time, Oxford, England, September 2011. work strain and employee distress. Paper presented at A Decade in Internet

memory: Effects of messaging interruptions on memory and performance. Paper interruption: Influence of task type on performance. Paper presented at the

OZCHI2000 Conference Proceedings: 356-361, Sydney: Ergonomic Society of

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS 86: 499-512.

Demerouti, Evangelica, Arnold B. Bakker, Friedhelm Nachreiner, and Wilmar B. Schaifeli.

Desjardins Financial Security (2006). Desjardins Financial Security of Health and the Health%20survey%202006%20Health%20is%20cool.pdf http://www.desjardinslifeinsurance.com/en/life-events/Documents/

2001. The job demands-resources model of burnout. Journal of Applied Psychology Desjardins National Financial Security Index. Retrieved September 10, 2011, from,

30

Duxbury, Linda, Ian Towers, Christopher Higgins, and John A. Thomas. 2006. From 9 to 5 to 24 and 7: How technology redefined the work day. Pp. 305-332 in Information Resources Management: Global Challenges, edited by W. Law. Hershey, PA: Idea Group Publishing. Lifestyle, Public Health Agency of Canada. Retrieved October 2010

Duxbury, Linda, and Christopher Higgins. 2001. National Study of Work, Family and (http://www.phacaspc.gc.ca/publicat/work-travail/index-eng.php).

Fenner, Grant H., and Robert W. Renn. 2004. Technology-assisted supplemental work: Fenner, Grant H., and Robert W. Renn. 2010. Technology-assisted supplemental work and Ferrer, Raquel Costa, Miguel Angel Serrano Rosa, Ana Zornoza Abad, and Alicia Salvador work-to-family conflict: The role of instrumentality beliefs, organizational expectations, and time management. Human Relations 63(1):63-82. construct definition and a research framework. Human Resource Management 43:179-200.

Fernandez-Montejo. 2010. Computer-mediated communication and time pressure

induce higher cardiovascular responses in the preparatory and execution phases of

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

Glavin, P., & Schieman, S. (2012). Work-family role blurring and work-family conflict: the moderating influence of jobs resources and job demands. Work and Occupations, 39(1), 71-98. Implications for training sponsorship, quit rates, and efficiency wages. Scottish Journal of Political Economy 47(3):251-72. 43(4), 709-741.

cooperative tasks. Psiothema 22(4): 697-702.

31

Green, Francis. 2000. The impact of company human resource policies on social skills:

Green, Francis. 2004a. Why has work effort become more intense? Industrial Relations, Green, Francis. 2004b. Work intensification, discretion and the decline in well-being at Groves, R. M. (2006). Nonresponse rates and nonresponse bias in household surveys. Public Opinion Quarterly, 70(5), 646-675. work. Eastern Economic Journal 30(4):615-624.

Hill, E. Jeffrey, Maria Ferris, and Vjollca Martinson. 2003. Does it matter where you work? Influences aspects of work and personal/family life. Journal of Vocational Behavior 63:220-241.

A comparison of how three venues (traditional office, virtual office, and home office)

Kalleberg, A. L., Marsden, P. V., Reynolds, J., & Knoke, D. (2006). Beyond Profit? Sectoral Kennedy, Tracy L. M., Aaron Smith, Amy Tracy Wells, and Barry Wellman. 2008. www.pewinternet.org).

differences in high-performance work practices. Work and Occupations, 33, 271-302.

Kurland, Nancy. B., and Diane E. Bailey. 1999. Telework: The advantages and challenges of

Networked Families, Pew Internet and American Life Project. Retrieved October 2010

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

Lawler, Edward. 1986. High-involvement management. Jossey-Bass: San Francisco, CA. Macky, Keith and Peter Boxall. 2008. High-involvement work processes, work experiences. Asia Pacific Journal of Human Resources 46(1): 38-55. Retrieved October 2010 (http://www.pewinternet.org). intensification and employee well-being: A study of New Zealand worker

working here, there, anywhere, and anytime. Organizational Dynamics 28:53-68.

32

Major, Virginia Smith, Katherine J. Klein, and Mark G. Ehrhart. 2002. Work time, work 87(3):427-436.

Madden, Mary and Sydney Jones. 2008. Networked Workers, Pew Research Center.

Mamaghani, Farrokh. 2006. Impact of information technology on the future workforce of distress.Computers in Human Behavior 26:61-69. Computing Systems (CHI 08), Florence, Italy. the future: An analysis. International Journal of Management 23(4):845-850.

interference with family, and psychological distress. Journal of Applied Psychology

Mark, Gloria, Daniela Gudith, and Ulrich Klocke. 2008. The cost of interrupted work: More Maume, David J. and David A. Purcell. 2007. The over-paced American: Recent trends managerial and professional work: Implications for managing the work-life in the intensification of work. Research in the Sociology of Work 17: 251-283. speed and stress. Presented at 26th annual SIGCHI conference on Human Factors in

Mano, Rita S. and Gustavo S. Mensch. 2010. E-mail characteristics, work performance and

Milliken, Frances J. and Linda M. Dunn-Jensen. 2005. The changing time demands of Individual Perspectives, edited by E. E. Kossek and S. J. Lambert. Mahway, NJ:

boundary. Pp. 43-59 in Work and Life Integration: Organizational, Cultural, and

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS Lawrence Erlbaum Associates.

Murray, William C. and Adam Rostis. 2007. Whos running the machine? A theoretical ODriscoll, Michael P., Biron, Caroline, and Cary L. Cooper. 2009. Work-Related University Press. Individual Employment Rights 12(3): 249-263. exploration of work stress and burnout of technologically tethered workers.

33

Thoits, Peggy A. 2010. Stress and health: Major finding and policy implications. Journal of Towers, Ian, Linda Duxbury, Christopher Higgins, and John Thomas. 2006. Time thieves United States Department of Labor Bureau of Labor Statistics. 2004. Work at Home In 2004. Washington, DC: US Department o Labor, available at: http://www.bls.gov/news.Release.homey.pdf Management 19(5):593-618. Health and Social Behavior 51(S):S41-S53. and space invaders: Technology, work and the organization. Organizational Change

Stiroh, Kevin, J. 2002. Information Technology and the U.S. Productivity Revival: What do the industry data say? The American Economic Review 92(5):1559-1576.

Psychological Well-being, edited by Y. Amichai-Hamburger. New York, NY: Cambridge

Technological Change and Psychological Well-being. Pp. 106-130 in Technology and

Valcour, P. Monique and Larry W. Hunter. 2005. Technology, organizations, and work-life integration. Pp. 61-84 in Work and Life Integration: Organizational, Cultural, and Individual Perspectives, edited by E. E. Kossek and S. J. Lambert. Mahway, NJ: Lawrence Erlbaum Associates.

Wajcman, Judy. 2008. Life in the fast lane? Towards a sociology of technology and time.

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS The British Journal of Sociology 59(1):59-77.

Wall, T. D. Michie, J., Patterson, M., & Stephen J. Wood. 2004. On the validity of subjective Walton, Richard. 1985. From control to commitment in the workplace. Harvard Business Review 63(2):77-84. measures of company performance. Personnel Psychology 57(1):95-118.

34

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

Noelle Chesley is an Assistant Professor in the Department of Sociology at the University of has been published in scholarly journals such as Journal of Marriage and Family, Gender & Society, and Information, Communication & Society. Britta Johnson has an M.A. in sociology from the University of Wisconsin-Milwaukee and employment.

AUTHOR BIOGRAPHIES

35

WisconsinMilwaukee. Her research examines the role of technological innovation, gender,

and race/ethnicity in shaping the work/family interface and individual well-being. Her work

works as an account executive for ADP in Milwaukee, WI. Ms. Johnson's research focuses on the sociology of work, the work-family intersection, and the influence of technology use on

Table 1: Descriptive Statistics (N = 712)

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

36

Variables in the Analysis %/Mean Std. Err. Dependent Variables % Increased Distress (1= "a lot" or "some") 0.33 % Increased Productivity (1 = "a lot" or "some") 0.69 ICT Based Work Practices Work Extension (low = 12 to high = 48) 21.83 0.37 % Increase in Telecommuting 0.18 % Decrease in Telecommuting 0.06 Network Expansion (1 = low to 4 = high) 2.99 0.05 % Arrange Meetings--In Person 0.14 % Arrange Meetings-- ICT 0.52 % Edit/Review Docs--In Person 0.21 % Edit/Review Docs--ICT 0.46 % Ask Questions--In Person 0.45 % Ask Questions--ICT 0.22 % Sensitive Issues--In Person 0.59 % Sensitive Issues--ICT 0.07 % Problems to Supervisor--In Person 0.54 % Problems to Supervisor--ICT 0.12 Total Work Tasks--In Person (0 to 5) 1.94 0.07 Total Work Tasks--ICT (0 to 5) 1.39 0.06 Control Variables Age (years; 18 - 88) 39.60 0.50 % female 0.47 % Non-Hispanic White 0.77 % No High School 0.07 % With High School Diploma 0.29 % Some College 0.29 % College/Advanced Degree 0.35 Average Weekly Work Hours (1 - 61) 40.82 0.54 % Income less than 30K 0.35 % Income 31 - 49K 0.27 % Income 50 - 74K 0.18 % Income 75K or greater 0.20 % in professional job 0.22 % in managerial/executive job 0.09 % in clerical or sales 0.20 % in other 0.31 % in service 0.18 % Self Employed 0.12 % with Large Private Employer 0.31 % with Medium Private Employer 0.16 % with Small Private Employer 0.27 % with Other (Govt/Non-Profit/Other) 0.26 % Reporting high job satisfaction 0.90 Job Autonomy (1 = low to 5 = high) 3.67 0.05 Job Advancement (1 = low to 5 = high) 3.35 0.05 Job Complexity (4 = low to 20 = high) 13.76 0.14 Source: PEW 2008 Networked Workers Study (weighted) Note: SEs are estimated using Taylor Linearization Method to account for design effects.

Table 2: Logistic Regression Results Illustrating the Influence of ICT Based Work Practices on Employee Distress and Productivity (N = 712) Work Extension Distress Produtivity Odds Odds Ratio Ratio SE SE Age Female White No HSa
HS Diploma a

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

Telecommuting Distress Productivity Odds Odds Ratio Ratio SE SE 1.01 1.22 0.98 1.89 0.64 0.97 1.02 0.82 0.89 0.71 1.56 1.60 2.42 1.10 1.13 0.83 0.50 0.61 0.33 1.02 1.07 1.03 2.24 1.10 0.01 0.26 0.24 0.91 0.19 0.26 0.01 * 0.28 0.26 0.21 0.57 0.64 0.87 * 0.39 0.36 0.22 0.17 * 0.17 0.11 ** 0.09 0.08 0.03 0.56 ** 0.46 1.00 1.30 1.05 0.49 0.46 0.62 1.01 1.12 1.52 1.48 1.40 2.02 1.86 0.80 1.08 1.43 1.54 1.22 1.47 0.97 1.02 1.07 6.89 1.98 0.01 0.30 0.28 0.22 0.13 ** 0.18 0.01 0.39 0.49 0.49 0.54 0.83 0.63 0.24 0.37 0.43 0.51 0.37 0.50 0.08 0.08 0.04 * 2.97 ** 1.11 1.43 0.14 *** 2.19 0.23 *** 1.20 1.10 0.07 ** 0.08 1.36 1.40 Communications Network Expansion Distress Productivity Odds Odds Ratio Ratio SE SE 1.01 1.20 0.99 1.53 0.57 0.82 1.03 0.93 0.92 0.66 1.57 1.55 2.19 1.14 1.29 0.94 0.56 0.62 0.31 1.02 1.05 1.03 0.01 0.25 0.24 0.77 0.16 0.21 0.01 ** 0.31 0.26 0.20 0.57 0.64 0.79 * 0.40 0.44 0.25 0.19 0.18 0.11 ** 0.09 0.08 0.03 1.00 1.27 1.11 0.35 0.42 0.49 1.02 1.55 1.84 1.35 1.40 2.04 1.42 0.89 1.26 1.90 2.07 1.27 1.41 0.95 0.95 1.08 0.01 0.31 0.32 0.18 * 0.13 ** 0.15 * 0.01 0.57 0.61 0.48 0.55 0.89 0.53 0.29 0.42 0.61 * 0.73 * 0.40 0.51 0.09 0.08 0.04 *

37
General Work Tasks Distress Productivity Odds Odds Ratio Ratio SE SE 1.01 1.30 0.97 1.87 0.60 0.90 1.03 0.82 0.84 0.62 1.44 1.32 2.15 1.01 1.31 1.00 0.57 0.66 0.31 1.03 1.05 1.03 0.01 0.28 0.24 0.94 0.18 0.24 0.01 ** 0.28 0.24 0.18 0.54 0.54 0.78 * 0.36 0.44 0.27 0.20 0.19 0.10 ** 0.09 0.08 0.03 1.00 1.47 1.09 0.60 0.61 0.71 1.01 1.23 1.45 1.22 1.23 1.19 1.29 0.72 1.40 1.94 1.98 1.38 1.31 1.00 1.00 1.07 0.01 0.37 0.31 0.31 0.20 0.20 0.01 0.46 0.47 0.42 0.51 0.52 0.47 0.23 0.49 0.60 * 0.67 * 0.42 0.45 0.09 0.09 0.04

1.01 0.01 1.37 0.29 0.99 0.24 1.73 0.90 0.60 0.18 0.92 0.24 1.02 0.01 * 0.90 0.32 0.97 0.29 0.74 0.23 1.45 0.53
c b

1.00 1.53 1.04 0.45 0.44 0.63 1.01 1.21 1.69 1.58 1.41 1.46 1.68 0.81 0.82 1.84 1.80 1.26 1.59 0.91 1.01 1.06 1.09

0.01 0.37 0.28 0.21 0.13 ** 0.18 0.01 0.42 0.55 0.54 0.56 0.61 0.58 0.25 0.29 0.57 * 0.60 0.39 0.54 0.08 0.08 0.04 0.02 ***

Some College a Work Hours Income Less than 30Kb Income 30-49K Professional
c

Income 50-74Kb Manager/Executive Clerical/Salesc Other Self Employed Large Private Employerd Medium Private Employer
d c

1.25 0.54 2.15 0.76 1.08 0.38 0.96 0.34 0.98 0.26 0.55 0.19 0.62 0.34 0.99 1.06 1.02 1.05 0.18 0.11 ** 0.08 0.08 0.03 0.01 ***

Small Private Employerd Job Satisfaction Job Autonomy Job Advancement Job Complexity Work Extension (Phone & Email) Telecommuting Increased Telecommuting Decreased Comm. Network Expansion Complete Tasks in Person Complete Tasks Using ICT

0.09 *** 0.13 ***

Source: PEW 2008 Networked Workers Survey; a Comparison is College/Advanced Degree; b Comparison is Income 75K or greater; c Comparison is Service Work; d Comparison is Public/Non-Profit/Other * p < 0.05; ** p < 0.01; *** p < 0.001

TECHNOLOGY USE, PRODUCTIVITY, AND DISTRESS

Table 3: Abbreviated Logistic Regression Results for Models Testing the Influence of Work Tasks on Distress and Productivity (N = 712) Distress Odds Ratio SE 2.28 2.03 2.87 1.68 2.14 2.01 2.14 1.85 2.07 2.24 0.73 * 0.52 ** 0.79 *** 0.44 * 0.54 ** 0.56 * 0.52 ** 0.77 0.51 ** 0.73 * Productivity Odds Ratio SE 2.94 6.63 5.21 4.75 5.38 4.03 5.27 3.42 4.86 5.43 0.91 ** 1.87 *** 1.59 *** 1.35 *** 1.38 *** 1.40 *** 1.36 *** 1.55 ** 1.28 *** 2.13 ***

38

Work Tasks a, b Arrange Meetings/Appointments In Person Using ICT Edit/Review Documents In Person Using ICT Ask Questions About Work In Person Using ICT Deal With Sensitive Issues In Person Using ICT Bring Problem To Supervisor In Person Using ICT

Source: PEW 2008 Networked Workers Study

a b

Comparison category in all cases is "Does Not Use ICT at Work"

All Models control for age, gender, race, education, income, hours worked, job type, organization type, job satisfaction, job autonomy, job advancement, and job complexity. * p < 0.05; ** p < 0.01; *** p < 0.001.