A Correlational Pilot Study Examining
Affect and Procrastination on the Internet
Kathy Harriott
Fifty-four participants, between 16 and 58 years old took part in this study examining the relationship between how much time people spend on the Internet, the degree of procrastination, and how they felt about their time on-line. Three questionnaires were posted on the World-Wide Web to collect data. The first questionnaire asked questions pertaining to demographics, and feelings connected to use of the Internet. The second was the Composite Measures of Affect (Diener & Emmons, 1985). The third was the General Procrastination Scale (Lay, 1986). As predicted, positive correlations were found between Internet procrastination and negative affect, and negative correlations were found between Internet procrastination and positive affect. However, when the sample was divided into student/non-student groups, students procrastinating on-line were not feeling significant negative affect. It is unclear if the Internet is causing procrastination or if the Internet is just a tool used to procrastinate. Although the sample was small the findings are interesting and further research is indicated to establish causality. iii
A Correlational Pilot Study Examining Affect and
Procrastination on the Internet
Kathy Harriott
A number of authors and researchers have noted the prevalence of Internet use and recognize the resultant social changes that are occurring in human relations because of it (e.g., Griffiths, 1997; Kraut, Patterson, Lundmark, Kiesler, Mukopadhyay & Scherlis, 1998; Smith & Leigh, 1997). These changes have been compared to those surrounding the arrival of the radio, the telephone and television (Griffiths, 1997; Kraut et al., 1998). The popular press (e.g. Psychology Today) has also become interested in Internet use and have published articles specifically discussing Internet use and overuse. On the whole, these authors and researchers agree that the Internet has, and will continue to have, an impact on our lives and consequently is an important area of continued research.
Computing Canada (Feb. 23, 1998) states that there are 30 million machines connected to the Internet, and that there had been a 14% growth over the previous 6 months. In the US, it is believed that 40% of households will have personal computers, and 1/3 of those will have access to the Internet (Kraut et al.,
1998). However, as Scheuermann and Langford (1997, p. 847) point out, "Every asset has the potential for abuse." Just as the telephone and typewriter were available to employees to conduct personal business, so is the Internet. Michael Welles, president of EdWel & Co., a performance and training consulting firm, declares that the Internet is the "water cooler" of the nineties (Sunoo, 1996), only better, because employees look as if they are busy.
If people are using the Internet as the high-tech water cooler, and if on-line access at work is increasing, then "loafing" on the job with use of the Internet is probably occurring. According to a survey conducted by Sunoo, (1996), 63% of Human Resource managers believe employees are spending increasing amounts of time "surfing the net" or engaging in other distracting past-times. This is a problem given that loafing on the job, by definition, is stealing from the employer. Sunoo (1996) calculates that if each employee of a large company (i.e., 1,000 employees or greater) wastes time on the Internet for 1 1/2 hours per day, at $20 per hour, the company is losing $30,000.00 per day to idleness. Nancy Probst of the North Broward Medical Centre in Florida suggests that if workers are loafing on the job
that it may not be intentional, but due to feelings of "fear,
depression and malaise" (cited in Sunoo, 1996). Sunoo argues that these feelings may be promoted and exaggerated by the strain of downsizing, having to learn new technologies, stress or tedium.
While some researchers believe that Internet use may be as benign as loafing, other researchers are exploring the possibility that use or overuse of the Internet may be more accurately described as an addiction (Griffiths, 1995; Shaffer 1996; Shotton 1991; Brenner, 1997 and Young, 1996). Griffiths (1995) established that technological addictions exist, while also attempting to develop criteria that could be applied to determine whether or not a person is addicted to Internet use. Shaffer (1996) also discusses Internet gambling and Internet use in terms of addictions. As he indicates, there is no shared definition of addiction in current diagnostic manuals, a viewpoint also held by Young (1996). Shaffer's contention is that addiction is the result of shifts in subjective experience, and that new technology and the Internet are dependable ways to alter subjective experience (Shaffer, 1996). According to Shaffer, it is the relationship a person has with the object of
their addiction, either behavioural or physical, that defines addiction and addictive activities. Young (1996) also holds this
belief, and also points out that the amount of time spent on the Internet should not be the only criteria used for assessing whether an addiction is present. She argues that other symptoms of Internet addiction, like alcohol and drug addiction, include tolerance, withdrawal, conflict and relapse (Young, 1996). Griffiths' (1996) research supports Shaffer's and Young's work and presents evidence that behavioural addictions have comparable consequences to the more researched and accepted alcohol and drug addictions. For instance, some comparable effects of behavioural and physical addictions are loss of sleep, marital breakdown, and neglecting social obligations and other commitments (Griffiths, 1996). Although there is growing acceptance that addictions can be behavioural, and Moulton (1998) describes Internet addicts as "Cyberdependents" or suffering from "Internet Addiction Disorder," the Addiction Research Foundation (ARF) states this classification is not universally recognized. Consequently, associating addiction with Internet overuse may be hasty (Brenner, 1997, and Huang and Alessi, 1997).
Given this reluctance of the ARF, defining Internet use as
an addiction may be overstating the issue of Internet overuse. Perhaps "loafing" on the Internet may be more aptly described as procrastination. In general terms, the popular press has taken
up the matter of procrastination, with publications on the topic in such forums as Seventeen, Homemaker's Magazine, and the Financial Post. Considering the diverse audience of these periodicals, it is clear procrastination is a topic of interest and concern to many people. Researchers who have delved into the area of procrastination, have used different definitions of procrastination, yet they have highlighted common themes throughout.
For example, Silver and Sabini (1981), argue that procrastination involves recognizing "what one ought to do." They contend that we choose a procrastinatory activity that can be completed in a short amount of time so that it can be dropped quickly and we can return to the task at hand. In this respect procrastinating on the Internet may fit Silver and Sabini's (1981) conception, as we often work in various capacities and for a variety of purposes at our computers, and the Internet can be quickly and easily accessed, yet not take us away from our work station. To others, it would appear we are still hard at work.
In addition, log-on times can be brief. After all, it only takes a minute or two to check our email.
Ferrari (1991a), on the other hand, describes both behavioural and decisional procrastination. The first being a
"general tendency to postpone everyday tasks" and the latter being an "inability to make timely decisions." Ferrari (1991b) contends that by delaying completion of a task, procrastinators are able to avoid situations where they will be evaluated and choose circumstances where their public image will not be damaged by a poor performance. Given Ferrari's assertion, one might expect high procrastinators to be drawn to the anonymity of the Internet where their public image can be protected.
It has also been suggested that procrastination is a form of revenge, and that retaliation may be occurring because of perceived wrongs (Kim, Smith & Rubin, (1991) as cited in Ferrari and Emmons, 1994). Procrastination as a "passive-aggressive form of revenge" may be used by those who feel powerless and have no other means of revenge (Ferrari & Emmons, 1994). Lay (1986), also suggests procrastination may be a type of rebellion against those in power based on rancour because of "perceived unfairness." If stress and downsizing are felt to be unjust by
employees, and is promoting "loafing" as Probst suggests (Sunoo, 1996), then procrastination on the Internet may be a form of revenge by employees.
Not all research, however, has found computer use to be linked with procrastination. For example, Walther's (1992)
research on Computer-Mediated Interaction (CMI) suggests that participants in CMI groups are more task-oriented than face-to-face groups. A few cautionary notes are warranted, however, in that Walther's study was not directly related to Internet use, and all of the experiments determining the task-oriented nature of participants in Walther's study were time limited, which may have affected outcomes. Although this study was not directly related to Internet use, there is reason to believe that company and school work will be, and is being, conducted via email and the Internet, and the Internet is currently being used to expedite work schedules and meetings. Walther found that there was less dominance by one individual or group, and there was a greater equality of participation and greater status equality in CMI group discussion in comparison to face-to-face group discussion.
Siegel, Dubrovsky, Kiesler & McGuire (1986) also found
greater participation in CMI groups with less dominance by one person and less inhibited interchanges measured by the use of "inflammatory" responses and "swearing and name calling." Siegel et al. (1986) believe computer mediated interaction will decrease "feelings of embarrassment," yield "less social comparison with others" as well as diminish "fears of penalty and
disapproval." In contrast to Walther's findings, Siegel et al. (1986) found group decisions using computer mediated interaction were slower than face-to-face interaction.
The Internet is a relatively new medium and consequently very little research has been done examining the social ramifications of Internet use. Research concerning Internet use has focused primarily on the perceived addictive qualities of the Internet and the links between procrastination and Internet use have not been studied. The present study was designed to explore if and how people use the Internet to procrastinate. This survey was devised to examine how much time participants spent on the Internet, how they felt about the time they spent on-line, and whether that time was productive, as opposed to procrastinatory. It was hypothesized that the more time participants spent unproductively on-line, the more negative affect they would feel. 8
Conversely, it was hypothesized that those who regard the Internet as entertaining and an important tool would experience less negative affect and greater positive affect.


The data were collected by means of a questionnaire posted on the Internet. To participate in this survey the only criterion was access to the Internet. The questionnaire could be found on the Internet by visiting the Procrastination Research Group (PRG) web page (Appendix A) at http://www.carleton.ca/~tpychyl. Participants were not actively recruited and could only have found the survey if they were looking for information on procrastination. It is therefore probably a safe assumption that those who accessed the PRG web page were self-identified procrastinators.
The survey posted to the World Wide Web consisted of four parts: 1) demographics (sex, age, occupation, citizenship, country of residence, and the State or Province if residing in Canada or the USA, respectively). 2) Questions pertaining to the individuals feelings and use of the Internet. 3) General Procrastination Scale (GP, Lay, 1986), and 4) Measures of Composite Affect (Diener & Emmons, 1985).
Feelings and Use of the Internet. Feelings and use of the
Internet was divided into two sections. The first part was a general measure of personal perceptions about the Internet and consisted of six items. For example, "I think the Internet is an important tool" and "I often use the Internet to relieve stress." Each item was scored on 5-point Likert-type scale (1=false of me; 5=true of me). The second part of this section was an extension of the first, the difference being qualitative rather than quantitative. This section consisted of five open-ended items, such as "How many hours do you spend on the Internet every day?" and "How many hours do you think you use the Internet to procrastinate?" Responses to the qualitative questions, "Why do you use the Internet?" and "How do you think about the time you spend on the Internet?" were used for anecdotal support and provided comparative and collaborative support for the quantitative findings.
General Procrastination Scale. The General Procrastination Scale (Lay, 1986) is a uni-dimensional measure of procrastination. This version of the GP consists of 20 items, for example, "I generally delay before starting on work I have to do" and "In preparing for some deadlines, I often waste time by
doing other things." Each item was scored on a 5-point Likert-
type scale (1=false of me; 5=true of me). Ten items were
reverse-scored and ratings were summed for a single-scale score. The GP has been found to have a Cronbach alpha coefficient of .78 (Ferrari, 1991) and test-retest reliability of .80 (Ferrari, 1989). Lay (1986) also found the GP to have a Cronbach alpha coefficient of .83. The GP Scale has been found to measure procrastination across a number of situations (Lay, 1986). For example, Lay found the scale to have predictive validity in a sample of travellers recruited at an airport. The GP scale was administered and the participants were then asked to return an envelope by post either on arrival at their destination, or three days later. It was found that those who scored higher on the GP scale were more likely to miss the assigned day for returning the envelope (Lay, 1986). Lay also reports the scale has significant correlations with rebelliousness (.34), organization (-.49), and neurotic disorganization (.69), and achievement (-.09), suggesting good concurrent validity (Lay, 1986).
Composite Measures of Affect. The Composite Measures of Affect (Diener & Emmons, 1985) consists of nine adjectives. Four of the adjectives represent positive affect items (i.e., happy,
joyful, pleased, and enjoyment/fun). The remaining five adjectives represent negative affect items (i.e., unhappy,
depressed, frustrated, worried/anxious and angry/hostile). Two
additional measures of affect were added to the scale, guilt and motivation. Guilt was included because there has been a demonstrated relationship between procrastination and guilt (Pychyl & Little, 1998). Motivation was added to the positive affect items because it has been found that motivation has a negative correlation with procrastination (Pychyl, Lee, Thibodeau and Blunt, in press). Participants indicated the extent to which they had felt each of the eleven moods on a 7-point Likert-type scale, ranging from 0 (not at all) to 7 (extremely much). A composite positive affect score was produced by summing the positive affect adjectives. The negative affect adjectives were also summed to produce a composite negative affect score. Reliability and internal alpha coefficients of the composite affect scales have been demonstrated to approach .90 in early studies (Diener & Larsen, 1984; Diener, Larsen, & Emmons, 1986), and do not correlate significantly with measures of social desirability (Emmons & Diener, 1985).

When the Procrastination Research Group web page was accessed, it offered a number of choices for those interested in procrastination, "Participate in Research" being one of the
choices (Appendix A). Once an interested individual clicked on
"Participate in Research," the informed consent was accessed (Appendix B). The informed consent expressed the voluntary nature of the questionnaire, explained confidentiality of responses and instructed participants that they could withdraw at any time. An explanation of what would be required in participating and the purpose of the research was also presented in the informed consent. This section provided details on how to receive further information about the project if required by participants. As it was not possible to collect signatures, submission of answers was considered informed consent.
A hypertext link, highlighted as "questionnaire," was displayed at the end of the informed consent. Those individuals wishing to continue could click on this link and the questionnaire (Appendix C) was presented. The first section presented was demographics followed by Feelings and Use of the Internet. The General Procrastination Scale was the next scale
presented and finally the Composite Measures of Affect scale was displayed. When the questionnaire was completed, participants could click on an icon labelled "send" or "reset". By choosing "send" the survey was submitted and choosing "reset" allowed participants to withdraw or return to the questionnaire to change
their responses.
Finally, a hypertext link was displayed for debriefing information, and once this link was accessed, the debriefing was displayed. The debriefing (Appendix D) included a reference list for further reading on procrastination and suggestions for self-help books were supplied.

Sixty surveys were received, however, six were discarded because of incomplete or inaccurate information. Two surveys were left incomplete and one survey did not specify the sex of the respondent and these were subsequently deleted. Three other surveys were discarded because of inconsistencies in the amount of time spent on the Internet. One of these reported spending 24 hours a day on the Internet and two reported spending zero hours per day on the Internet. Therefore 54 surveys remained, 32 female respondents and 22 male.
The average age was 34.66 (SD 12.16). The youngest age represented was 16 and the oldest 58. Although the sample size in this study was small, and therefore biased, the mean age was consistent with other studies examining Internet and computer use. For example, the age range in Shotton's (1989) study was 14-64, mean age 29.7. Young's (1998) participants mean age was 31.5, while Computing Canada (Solomon, 1998) reports the average age of Internet users as 31. Brenner (1997), Shaffer (1996) and

Griffiths (1997) research did not report the age of their participants.
Participants reported five countries of citizenship and seven countries of residence. The majority were US citizens (26); Canadian citizens were the second most frequent respondents (23); two participants were Australian citizens, one Mexican citizen, one New Zealand citizen, and one survey left the citizenship question unanswered. The countries represented were, Canada, USA, Australia, Germany, Bermuda, Italy and one survey from New Zealand which was one of the surveys finally deleted. Although participants were asked to give the region they lived in, if they were responding from Canada or the US, these data
were not analysed because the sample was small and not enough variability could be provided. The participants occupations were differentiated only as either "student" (N=21) or "non-student" (N=33). Students mean age was 24.1 (SD 6.11), with one case missing, and the non-student group mean age was 41.06 (SD 10.32).
Feelings and Use of the Internet
Responses to questions one to six in Section 1 covered the whole range of possible responses, that is "False of me" to "True of me", except for two statements (Table 1). Not one participant
Table 1
Feelings and Use of the Internet

Q 1
I think the Internet is Frequency Percent
an important tool. S Non-S S Non-S

False of me 1 1 4.8 3.0
Sometimes true\false of me 4 2 19.0 6.1

True of me 16 30 76.2 90.9

Q 2
I think the Internet is Frequency Percent
very entertaining S Non-S S Non-S

False of me 2 2 9.6 6.0
Sometimes true\false of me 3 8 14.3 24.2

True of me 16 23 76.2 69.7

Q 3
I feel the Internet makes me Frequency Percent
more productive S Non-S S Non-S

False of me 6 3 28.6 9.1
Sometimes true\false of me 10 14 47.6 42.4

True of me 5 16 23.8 48.5


Table 1 cont.

Q 4
When I am on line,
I end up spending more time Frequency Percent
than I anticipated S Non-S S Non-S

False of me 2 5 9.5 15.2

Sometimes true\false of me 2 15 9.5 45.5
True of me 17 13 81.0 39.4

Q 5
I often use the Internet Frequency Percent
to relieve stress S Non-S S Non-S

False of me 10 10 47.6 30.3

Sometimes true\false of me 3 8 14.3 24.2

True of me 8 15 38.1 45.4

I often use the Internet Frequency Percent
to procrastinate S Non-S S Non-S

False of me 5 8 23.8 24.3
Sometimes true\false of me 3 8 14.3 24.2

True of me 13 17 61.9 51.5
Note: S = Student Non-S = Non-student

responded "False of me" to the statement "I think the Internet is an important tool" and "When I am on line, I end up spending more time than anticipated". Table 1 represents frequencies and
percentage of responses for students and non-students, collapsed across groups. As can be seen in Table 1, most of the participants consider the Internet to be an important tool, very entertaining, spend more time on-line than anticipated, and often use the Internet to procrastinate. A small majority of non-students reported using the Internet to relieve stress, however, the majority of students reported they did not use the Internet to relieve stress. Overall, responses to participants feelings about the Internet making them more productive, was largely in the "sometimes true/false of me" category. However, when the sample was divided into the two groups, it became apparent that a small majority of non-students believed the Internet made them more productive and reported "true of me."
Questions 7 and 8 in Section 2 asked participants how much time they spent on the Internet at home, work and school, and how much of that time they believed to be procrastinatory. The total mean time on the Internet at home work and school was 3.35 hours (SD 3.1). The modal time spent on the Internet was two hours,
with a maximum of 17 hours. Table 2 displays the mean time spent on the Internet and the time spent procrastinating on the Internet in each domain of home, work and school. Table 2 also presents the proportion of time on the Internet that the
participants believed to be procrastinatory. The total mean time spent procrastinating on the Internet at home, work and school each day was 1.55 hours (SD 1.66). The modal time spent procrastinating on the Internet was 1 hour, with a maximum of 10 hours. The Internet procrastination survey (IPS) requested time spent on line in hours per day, whereas other studies collected this information in hours per week or month. Table 3 presents mean hours per day spent on the Internet for each survey where these data were available. Participants in each survey spent similar amounts of time on the Internet, with the exception of Young (1998) and Shotton's (1991) "dependents", suggesting the current survey is consistent with past research.
Recurring themes were noted amongst the participants for both questions 9 & 10, "Why do you use the Internet?" and "How do you think about the time you spend on the Internet?" Each set of responses could be broken down into approximately 15 sub-categories with one or two responses being the most prevalent.
Table 2
Time spent on the Internet

Proportion Range Mean of time in
(N=54) (SD) Procrastinating hours

Time spent on
the Internet 1.69 (1.95) 0-10

Time spent
on the Internet .97 (1.56) .57 0-10

Time spent on
the Internet 1.21 (2.46) 0-12

Time spent
on the Internet .32 (.68) .26 0-4

Time spent on
the Internet .45 (1.34) 0-8

Time spent
on the Internet .26 (.88) .57 0-4.5


Table 3

Comparative Time Spent on the Internet Between Studies

Study Hrs/day

IPS (work & home) 2.90

IPS (school & home) 2.14

Solomon 1.43

Brenner 2.71

Young (non-Dependents) .70

Shotton (Owners) 3.00

Young (Dependents) 5.50

Shotton (Dependants) 7.51

For instance, the responses to "Why do you use the Internet?" were mostly research/information and work/homework related, with 76% of the students and 88% of non-students reporting these activities as at least one of their reasons for using the Internet. In addition, 43% of all participants reported one of
the purposes of their Internet use to be either email, personal communication or Multi-User Dungeons (MUDS). Responses to "How do you think about the time you spend on the Internet?" were not quantified but were found to be equally divided between "wasted/unproductive" and "worthwhile/productive." Other responses to this question were quite scattered however, and were equally divided between positive and negative answers. For example, "harmless hobby," "relaxing," "healthy fun," and "guilty," "poorly," and "trying to reduce use."
Question 11, "Do you think you are procrastinating now?" received responses from "no" to "perhaps" to "yes," including "yes" with added emphasis (e.g. hell yes and absolutely). The majority responded "yes" (68.6%), while 24.1% said "no" and 7.4% said "perhaps".
Feelings and Use of the Internet and Positive and Negative Affect

Correlations were calculated between all of the questions in
Section 1 (e.g., "I think the Internet is an important tool" with "I think the Internet is very entertaining") for the two groups "student" and "non-student," and a number of significant correlations were found. "I think the Internet is an important tool" had a positive correlation with "I feel the Internet makes me more productive" for both students and non-students (r=.58,
<.01, and r=.44, p<.01, respectively). However, only the non-
students had a significant correlation between "I think the
Internet is an important tool" and "I think the Internet is very entertaining" (r=.46, p<.01), and "I often use the Internet to relieve stress" (r=.35, p<.05). Significant positive correlations were found between "I think the Internet is very entertaining" and "I often use the Internet to procrastinate" for both groups (r=.46, p.<05 for students, and r=.36, p<.05 for non- students). Only the student group had significant positive correlations between "I think the Internet is very entertaining" and "I often use the Internet to relieve stress" (r=.56, p<.01). The non-student group alone had a significant positive correlation between "I think the Internet is very entertaining" and "I think the Internet makes me more productive" (r=.50, p<.01). Students had a significant negative correlation between
"I feel the Internet makes me more productive" and "I often use the Internet to procrastinate" (r=-.59, p<.01). A significant correlation was found between "When I am on line, I end up spending more time than I anticipated" and "I often use the Internet to relieve stress" for non-students only (r=.44, p<.01). Finally, "I often use the Internet to relieve stress" correlated with "I often use the Internet to procrastinate" for both
students and non-students (r=.71, p.<001, and r=.78, p<.001, respectively).
A correlation coefficient was calculated between mean positive affect and questions one to six (e.g., I think the Internet is an important tool), and overall mean negative affect and questions one to six. It was found that mean positive affect correlated significantly with the first three statements (i.e., Important tool r=.44, p<.001; Entertaining r=.37, p<.01; and Productive r=.46, p<.001. Mean negative affect correlated with the last three statements (i.e., More time than anticipated r=.31, p<.05; Relieve stress r=.30, p<.05; and I often use the Internet to procrastinate r=.35, p<.01).
Table 4 presents correlation coefficients for negative affect and participants' feelings about their use of the
Table 4

Correlations Between Negative Affect and Internet Use

Variable Group Frustrated Worry Angry Depressed Guilt Unhappy

Important Both -.50*** -.33*
Student -.59** -.45*
N-student -.40*
Enter- Both -.37**
Student -.50*

N-student -.44**

Produc- Both -.30**
Student -.47*


More than Both .28* .30** .30**
ated Student .47* .49* .45*

Relieve Both .38** .50***

N-student .42** .55** .35**

Procrast- Both .29** .35** .51***

N-student .38* .50**

Note: N-student = Non-student *p<.05 **p<.01 ***p<.001

Internet. These correlations are presented for each group, student and non-student, as well as collapsed across all categories. The data suggest that those participants, who believe the Internet is an important tool, entertaining, and made them more productive, were experiencing less frustration (r=-.51, p<.001; r=-37, p<.01; and r=-.30, p<.05 respectively). Those in the group as a whole, who were attempting to relieve stress and procrastinating on the Internet, appeared to be experiencing greater feelings of worry, depression and guilt (r=38, p<.01 and r=.50, p<.001). Those procrastinating on the Internet were
apparently experiencing greater feelings of worry (r=.29, p<.01), depression (r=.35, p<.01) and guilt (r=.51, p<.001). However, when the group was divided into the sub-categories, only non-students appeared to be experiencing these feelings of negative affect. Those in the whole sample who were spending more time than anticipated on-line not only appeared to be encountering feelings of worry (r=.30, p<.01) and guilt (r=.30, p<.01), but increased levels of frustration (r=.28, p<.05). Finally, only the
student group was affected by negative affect when reporting they spent more time than anticipated on-line (Frustrated r=.47, p<.05; Worry r=.49, p<.05; and Guilt (r=.45, p<.01)).
Table 5 displays the correlation coefficients for positive affect and respondents feelings about their use of the Internet. Those participants who felt the Internet was an important tool, and felt it increased their productivity, appeared to be experiencing significant levels of positive affect in all of the positive affect categories presented. However, it became apparent that the students were experiencing greater positive affect for the "Important tool" category (Motivated r=.55, p<.01; Happy r=.60, p<.05; Joyful r=.46, p<.05; Pleased r=.56, p<.01). Participants who found the Internet entertaining were also
experiencing significant levels of enjoyment (r=.38, p<.01), happiness (r=.53, p<.001) and were feeling pleased (r=.44, p<.001). Both students and non-students were feeling happy and pleased while non-students only were feeling enjoyment. In the "productive" section, both students and non-students were feeling pleased but were not experiencing the same types of positive affect. The students were experiencing higher levels of motivation (r=.60, p<.01) and joyfulness (r=.66, p<.001), while non-students were experiencing greater enjoyment (r=.40, p<.05)
and happiness (r=.42, p<.05). The two statements "I end up spending more time on the Internet than I anticipated," and "I
Table 5

Correlations Between Positive Affect and Internet Use

Variable Group Enjoyment Motivated Happy Joyful Pleased
Important Both .31** .36** .46*** .27* .40**
Student .55** .60* .46* .56**

Entertain- Both .38** .53*** .44***
Student .65*** .46*

N-student .49** .44** .43**

Productive Both .33** .36** .28* .45*** .44***

Student .60** .66*** .46*
N-student .40* .42* .50**

More than Both
pated Student


Relieve Both


Procrast- Both -.28* -.27*
Student -.52*

N-student -.38

Note: N-sudent = Non-student *p<.05 **p<.01 ***p<.001

often use the Internet to relieve stress" had no significant correlations with any of the positive affect categories. Finally, significant negative correlations for motivation (r=- .28, p<.05) and joyfulness (r=-.27, p<.05) were related to using the Internet to procrastinate.
Correlation coefficients were also calculated between overall positive affect and general procrastination, and overall negative affect and general procrastination. The GP total and negative affect had a positive correlation (r=.43, p<.001), and the GP total and positive affect had a negative correlation
(r=-.29, p<.05). Guilt and motivation were correlated as individual items with the procrastination measure. Guilt was found to correlate positively with procrastination (r=.48, p<.01) and motivation correlated negatively with procrastination (r=- .41, p<.01).
Correlation coefficients between general procrastination and all of the negative affect categories, and general procrastination and the positive affect categories for the two groups "student" and "non-student" were calculated. Those who were classified as non-student had significant positive correlations with general procrastination and all of the negative
affect categories: frustrated (r=.46, p<.01), worried (r=.48, p<.01), angry (r=.44, p<.01), unhappy (r=.62, p<.001), depressed
(r=.50, p<.01), guilt (r=.68, p<.001). In addition, significant negative correlations for the non-students were found between general procrastination and motivation (r=-.58, p<.001) and general procrastination and happiness (r=-.58, p<.01). The students did not have significant correlations with either positive or negative affect categories and general procrastination.

Discussion In preface to the discussion of the results, it is important to present some caveats related to the data. There are some inherent problems with Internet research. First, as Smith and Leigh (1997) point out, there is no guarantee that results from Internet surveys are valid or generalizeable because participants may have a vested interest in participating in research. For example, in the current survey, five respondents clearly stated they were looking for help with their own procrastination problem. As Internet research participants are self-selected, this may yield a biased sample, however, it is expected participants are more likely to maintain their interest in the research until completion (Hewson, Laurent, & Vogel, 1996). Secondly, a large portion of the population, those who are unable to access the new technology, were not studied. Not only are the few who can afford to be a part of the new technology taking part in Internet surveys, but they may be further identified as elite members of a computer-literate public (Smith and Leigh, 1997). Finally, maintaining the integrity of
responses may be a problem. In this survey the anonymity of
respondents was jeopardized because email addresses were attached
to the survey responses. However, this also guarded against multiple responses from the same participant. It may be possible to avoid this problem by issuing a password to each participant before they are allowed to continue with the survey, thereby, protecting their anonymity and also preventing multiple replies (Smith and Leigh, 1997).
There are, however, a number of advantages for Internet research. One such benefit is that Internet surveys are believed to be resistant to experimenter effects and bias (Hewson et al., 1996). Secondly, using the Internet to do research increases the number of possible participants. Universities that have traditionally relied on students for their subject pool can now reach outside the campus boundaries, this is especially helpful to smaller universities which have a limited number of students (Smith & Leigh, 1997). Finally, possibilities for cross cultural studies are greatly enhanced and are more viable with Internet research, for example, the current survey attracted participants from seven countries. Despite the inherent limitations of Internet research, the data from this study are worthwhile
because a study of this nature has not been undertaken before and it may be the launching point for the development of more tightly
controlled Internet procrastination surveys and add to our
understanding of how future Internet procrastination research could be designed and conducted.
It will be remembered that the hypotheses driving this research was that those participants who believed that they were procrastinating on-line would experience more negative affect and those who felt the Internet is an important tool, their time spent on-line was productive, and entertaining, would feel more positive affect. The statistical analysis did in fact support this hypothesis for the sample as a whole (see Tables 4 & 5). As the participants belief that the Internet was an important tool increased, it made them more productive, and it was more entertaining, their frustration levels decreased, and enjoyment, happiness, and feeling pleased increased. Also, as procrastination on-line increased, the participants' experience of negative affect increased, most notably guilt, but also depression and to a lesser extent worry.
Perhaps one of the more interesting findings of this survey was that participants classified as students had no significant
correlations between any of the negative affect categories and
using the Internet to procrastinate (see Table 4 & 5). However,
the participants classified as non-student had a significant negative correlation between Internet procrastination and
motivation and significant positive correlations between worried and guilty and Internet procrastination. It is possible that the age difference between the two groups affected this outcome, the mean age for students was 24.1 years and for non-students was 41.06 years. This group was on average older, mostly in the work force, and possibly had more responsibilities. Perhaps the non-student groups' guilt and negative affect was being fuelled by other obligations. Given the advent of downsizing and the competitive nature of the current work force, if people are procrastinating at work on the Internet worry and guilt would appear to be an understandable outcome.
If the Internet is the water cooler of the nineties, as proposed in the introduction, should Human Resource managers be concerned? Or, can the water cooler effect be beneficial to business? The participants in this study believed they procrastinated on-line at work an average of .32 hours per day. Using Sunoo's (1996) criteria (i.e., 1000 employees x .32 x
$20/hr) this would cost the company $6400.00 per day, a considerable amount of money. However, only 33.3 % of the
participants in this research reported procrastinating on-line at all during working hours. Therefore, even if they all worked for the same company, the cost to the business would be $2112.00 per
day (i.e., 33% of 1000 employees = 330 x .32 x $20/hr). One would expect not only gossip at the water cooler, but an exchange of ideas and information, and for a large or small company this may just be the cost of doing business and maintaining a well informed staff.
Given Ferrari's (1991b) belief that procrastinators prefer to avoid evaluation and choose conditions where their public image will not be damaged, and pairing that with Siegel et al. (1986) contention that CMI reduces embarrassment and produces fewer social comparisons, one might surmise that higher procrastinators are drawn to the anonymity of the Internet. However, it is unclear whether those who are procrastinating on the Internet have found it to be just one more way to procrastinate, or if the Internet is encouraging procrastination. Causality will be difficult to establish, and the question remains, are high procrastinators drawn to the Internet, or is
the Internet a breeding ground for procrastinators?
According to Shaffer (1996) "Using a computer, like
narcotics, often accelerates the subjective passage of time" (p. 468). In response to the statement, "I often spend more time than anticipated on the Internet" not one participant answered "false of me," and a large percentage (81.5%) of participants
reported they often spend more time than anticipated while on the Internet. On the surface this may appear to be a function of addiction, however, the question remains, what is keeping participants on-line longer than they anticipated? There could be a number of reasons, and participants statements gave some hint as to the causes. For example, participants made such comments as [the Internet is] "frustratingly slow," "links are ambiguous," and "I am often led astray by detours." Down-loading information from the Internet can also be time consuming and that may be contributing to the amount of time spent on-line as well as increasing frustration levels, as at least one participant clearly articulated. Participants also expressed more positive experiences on-line which might also explain lengthy log-on times. For instance, some users are finding so much interesting information that they perceive their time on-line as well spent.
For example, some comments made by respondents were: "There are always interesting links to follow," and "Subjects which I might
not intensely research otherwise I explore ad infinitum," and simply, "I love it, I can't get enough of it." Although Internet users may be spending more time than they had planned, and it may be due to an accelerated passage of time, further inquiry is needed to clarify more precisely why people are spending so much
time on-line.
Some research supports that television watching has been used to reduce stress and to relax (Griffiths, 1995) and the Internet may "privatize entertainment," like the television (Kraut et al., 1998). Relaxing and reducing stress were reasons cited by participants in the current survey as reasons for using the Internet, much like television. However, using the Internet to relieve stress appears to be increasing levels of depression, unhappiness and guilt in the non-student group and positive affect is not being increased. Therefore, one might expect that procrastination on-line, at home especially, was diverting them from other more important tasks, perhaps child-rearing and family obligations, thus adding to their levels of negative affect.
It is important to note that five people expressly stated
they were looking for help with their procrastination and it is
expected that many of the other participants came to this web
page and survey for the same reason. Coupled with this, the demand for information about procrastination, as evidenced by articles in the popular press and professional journals, suggests there is a demand for an effective treatment plan. Given this need, is the Internet an appropriate forum for treatment? The idea of treatment delivery for various psychological problems via
the Internet has been proposed by Young (1998), Huang and Alessi (1997), and others. They also express the belief that with the phenomenal expansion of the Internet we can expect the possibility of entirely new disorders and new methods of treatment delivery. However, treating Internet "addiction" with the very tool that is the addictive "substance" would appear to be contraindicated. By the same token, if Internet overuse can be linked to procrastination, then the "cure" for the sufferer could be as dangerous as the "disease." Future research will need to focus on the matter of causality before an informed decision can be made about the appropriateness of on-line treatment for Internet procrastination.


Brenner, Viktor (1997). Psychology of computer use: XLVII. Parameters of Internet use, abuse and addiction: The first 90 days of the Internet usage survey. Psychological Reports, 80, 879-882.
Diener, Ed, & Larsen, Randy J. (1984). Temporal stability and cross-situational consistency of affective, behavioral, and cognitive responses. Journal of Personality and Social Psychology, 47 (4), 871-883.
Diener, Ed, Larsen, Randy J., & Emmons, Robert A. (1984) Person x situation interactions: Choice of situations and congruence response models. Journal of Personality and Social Psychology, 47 (3), 580-592.
Emmons, Robert A., & Diener, Ed (1985). Personality correlates of subjective well-being. Personality and Social Psychology Bulletin, 11 (1), 89-97.
Ferrari, Joseph, R. (1991a). Compulsive procrastination: Some self-reported characteristics. Psychological Reports, 68, 455-458.
Ferrari, Joseph, R. (1991b). Procrastination and project
creation: Choosing easy, non diagnostic items to avoid self-
relevant information. Journal of Social Behavior, 6 (3), 619- 628.
Ferrari, Joseph, R. (1992). Psychometric validation of two procrastination inventories for adults: Arousal and avoidance measures. Journal of Psychopathology and Behavioral Assessment, 14 (2), 97-110.
Ferrari, Joseph, R., Johnson, Judith L., McCown, William G. (1995). Procrastination and task avoidance: Theory, research, and treatment. Plenum Press, New York and London.
Griffiths, Mark (1995). Technological addictions. Clinical Psychology Forum, 17, 14-19.
Griffiths, Mark (1996). Internet "addiction": an issue for clinical psychology? Clinical Psychology Forum, 97, 32-36.
Griffiths, Mark (1997). Psychology of computer use: XLIII. Some comments on 'addictive use of the Internet' by Young. Psychological Reports, 80, 81-82.
Hewson, Claire M., Laurent, Dianna, & Vogel, Carl M. (1996). Proper Methodologies for psychological and sociological studies conducted via the Internet. Behaviour Research Methods, Instruments, & Computers, 28, 186-191.
Huang, Milton P., Alessi, Norman E. (1997). Drs. Huang and Alessi reply. American Journal of Psychiatry, 153 (6), 890.
Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T., & Scherlis W. (1998). Internet paradox. American Psychologist, 53 (9), 1017-1031.
Lay, Clarry H. (1986). At last, my research article on procrastination. Journal of Research in Personality, 20, 474- 495.
Moulton, Donalee (1998). Internet overuse mimics addictions. The Journal: Addictions News for Professionals, (ARF) 27 (1), 4.
Pychyl, T. A., Lee, J.M., Thibodeau, R., Blunt, A. (1998). Five days of emotion: An experience sampling study of undergraduate students. In Press Journal of Social Behavior and Personality.
Pychyl, T. A., & Little, B. R. (1998). Dimensional specificity in the prediction of subjective well-being: Personal projects in pursuit of the Ph.D. Social Indicators Research, 45.
Scheuermann, Lawrence E., & Langford, Harold (1997). Perceptions of Internet abuse, liability, and fair use. Perceptual and Motor Skills, 85, 847-850.
Shaffer, Howard J. (1996). Understanding the means and objects of addiction: Technology, the Internet, and gambling. Journal of Gambling Studies, 12 (4), 461-469.
Shotton, Margaret A. (1991). The costs and benefits of 'computer addiction.' Behaviour and Information Technology, 10 (3), 219-230.
Siegel, J., Dubrovsky, V., Kiesler, S., & McGuire, T.W. (1986). Group processes in computer-mediated communication. Organization Behaviour and Human Decision Processes, 37, 157-187.
Silver, Maury, & Sabini, John (1981). Procrastinating. Journal for the Theory of Social Behaviour, 11, 207-221.
Smith, Michael A., & Leigh, Brant (1997). Virtual subjects: Using the Internet as an alternative source of subjects and research environment. Behaviour Research Methods, Instruments, & Computers, 29, 496-505.
Solomon, Howard (1998). Facts, figures abound in new Internet study. Computing Canada, 24 (9), 13.
Sunoo, Brenda Paik (1996). This employee may be loafing can you tell? Should you care? Personnel Journal, 75, (12), 54-62.
Survey reveals 30 million machines on Internet. (1998). Computing Canada, 24 (6), 35.
Walther, Joseph B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19 (1), 52-90.
Young, Kimberly S. (1996). Psychology of computer use: XL.Addictive use of the Internet: A case that breaks the stereotype. Psychological Reports, 79, 899-902.
Young, Kimberly S. (1998). Internet addiction: The emergence of a new clinical disorder. In The Centre for On-Line Addiction [On-line], Available: http://www.netaddiction.com.