Computer anxiety and anger: the impact of computer use, computer experience, and self-efficacy beliefs
Introduction
Previous researchers have suggested that computer anxiety affects one-fourth of the population (Gos, 1996). Another researcher (Brosnan, 1998) has found that one-third of individuals among different populations experience some level of computer anxiety, varying from avoiding computers at all costs to minor stress. However, Rosen and Maguire (1990) have suggested that almost 50% of individuals display some sort of anxious behavior while using a computer. Whatever the exact percentage, it seems as though computer anxiety affects a great number of the people, and therefore must be addressed as a serious problem. Some researchers have implied that computer anxiety symptoms will abate with time resulting from a general increase of societal exposure to technology, however Bozionelos (2001b) has not found such a relation.
Computer anxiety is defined as a negative emotional state and/or negative cognition experienced by a person when he/she is using a computer or imagining future computer use (Bozionelos, 2001a). Computer anxiety is “state anxiety” which occurs at the time of computer use or at the time of imagined future computer use, and not inherently a personality trait (Chua et al., 1999, Cambre and Cook, 1985). Therefore, the computer anxiety symptoms of individuals should be measured only at these times.
Computer anxious individuals exhibit phobia-like symptoms which lead them to use computers less, and when using computers to complete tasks, they do so more slowly (Mahar, Henderson, & Deane, 1997). In addition, Bozionelos (2001b) suggested that computer anxiety causes decreased levels in psychological well-being. He further explained that if society continues to force computer technology onto computer anxious individuals, then this could actually result in a worsening of the already present anxiety symptoms. This suggests that systematic desensitization, a technique used in psychological learning theory, would help already anxious individuals approach and use computers in a systematic and gradual way after learning to remain calm.
Researchers have looked at possible factors which are correlated with computer anxiety including: specific computer experiences and domains (e.g. Bozionelos, 2001b, Chua et al., 1999, Gos, 1996), computer use (Bozionelos, 2003), and self-efficacy beliefs and outcome expectancies (e.g. Bandura, 1977, Bandura, 1986, Compeau and Higgins, 1995, Hill et al., 1987).
Traditionally researchers have used psychological measurement scales to operationalize computer anxiety (e.g. Hudiburg, 1990, Bozionelos, 2001a, Bozionelos, 2001b, Cambre and Cook, 1985, Rosen and Maguire, 1990). Cambre and Cook (1985) explained that many assessments tests have been developed and used, but as computer anxiety is believed to be a state anxiety, the most consistent and reliable device is the State-Trait Anxiety Inventory (Spielberger, 1983). The main approach is to modify the State-Trait Anxiety Inventory into Likert scale formats (Cambre & Cook, 1985). Other researchers agree with these guidelines (Mahar et al., 1997).
No research was found examining or explaining anger resulting from computer use. Anger is usually said to be the cognitive component of aggression, a larger term which also encompasses physical acts such as violence (i.e., the outward symptoms). “Anger consists of strong feelings of displeasure in response to a perceived injury” (Brehm, Kassin, & Fein, 2002). So, computer-anger is hereby defined as strong feelings of displeasure and negative cognitions in response to a perceived failure to perform a computer task. Anger is sometimes related with frustration. Brehm et al. (2002) explain that frustration is only one of many possible unpleasant experiences that can lead to aggression, including anger, by way of negative feelings.
Like anxiety, anger needs some sort of reliable and consistent measurement device. Adapting the State-Trait Anger Expression Inventory (Spielberger, 1988) into Likert scales would be a desirable choice. It is proposed that the same correlates affect both computer anxiety and computer anger, and these will now be discussed.
A measure of computer use would simply be to use the number of hours an individual used a computer per a week. Various researchers (e.g. Bozionelos, 2003, Bohlin and Hunt, 1995) have used computer use as either an antecedent or consequence of computer anxiety. Bozionelos (2003) showed that increased levels of computer anxiety correlated with decreased levels of computer use. In the present study the following hypotheses were used:
H1. Computer use will have a negative correlation with computer anxiety.
H2. Computer use will have a negative correlation with computer anger.
Bandura, 1977, Bandura, 1986 defined self-efficacy as “people’s judgments of their capabilities to organize and execute courses of action required to attain designated types of performances.” He also suggested that efficacy judgments vary according to three different measures: level (difficulty of task), generality (in regards to certain domains or situations), and strength (the power of beliefs which describe coping techniques). Self-efficacy is used as an important measure for fearful and avoidant behavior (Bandura, 1986). Bandura, 1977, Bandura, 1986 suggested that decreased levels of self-efficacy is related with increased levels of anxiety, which in turn is related with decreased levels of exposure to the stimulus (computer use). He explained that perceived competency might be compromised by stressful situations such as computer use accompanied by computer anxiety. Bandura (1986) stated that anxiety must be reduced in order for an individual to gain strength in efficacy belief. Social cognitive theory proposes, “it is mainly perceived inefficacy to cope with potentially aversive events that makes them fearsome” (Bandura, 1977). Thus, it would seem essential for any study of computer anxiety to have measures of self-efficacy as a correlate of computer anxiety, computer anger, or any other negative emotional state which contains cognitions. Bozionelos (2001b) also suggests that self-efficacy should be measured to predict anxiety.
An attempt to develop a reliable measure of self-efficacy beliefs was carried out by Compeau and Higgins (1995). Compeau and Higgins (1995) defined computer self-efficacy as representing “an individual’s perceptions of his or her ability to use computers in the accomplishment of a task (i.e., using a software package for data analysis, writing a mailmerge letter using a word processor), rather than reflecting simple component skills (i.e., formatting diskettes, booting up a computer, using a specific software feature such as ‘bolding text’ or ‘changing margins’).” In the spirit of Bandura, 1977, Bandura, 1986, the researchers developed a measurement device which took into account the “magnitude”, “strength”, and “generality” of self-efficacy. They were able to support much of Bandura’s hypotheses for self-efficacy in relation to computer use. The researchers explained that their theoretical model for self-efficacy was based on the social cognitive theory of “Triadic Reciprocality” (Bandura, 1977, Bandura, 1986), in which people, their behavior, and their environment are all intimately interconnected and interdependent. The following hypotheses were used in the present study:
H3. Computer self-efficacy will have a negative correlation with computer anxiety.
H4. Computer self-efficacy will have a negative correlation with computer anger.
Most researchers have found that computer experience is the most consistent correlate of computer anxiety (Chua et al., 1999, Brosnan, 1998, Mahar et al., 1997, Gos, 1996). Computer experience has been traditionally operationalized through self-report devices using point-scales which assess the mastery level in specific realms of computer software such as: word processing, programming, operating systems, etc. (eg. Hasan, 2003, Chua et al., 1999, Bozionelos, 2001a). Bozionelos, 2001a, Bozionelos, 2001b suggested that computer experience is the best measure for anxiety based on previous research. However, Gos (1996) suggested that the quality and/or pleasantness of prior computer experiences are a better measure for computer anxiety.
Computer experience is difficult for people to self-report as they are without a reliable reference point for comparison. They do not know what an expert is, nor do they know what a novice is. Possibly, people with already strong computer self-efficacy would rate themselves as more experienced than they actually are which would be consistent with Bandura, 1977, Bandura, 1986 social cognitive theory. Similarly, people with already weak computer self-efficacy could rate themselves as less experienced than they actually are.
Computer experience has also been used as a correlate for computer self-efficacy beliefs. Hasan (2003) shows a significant relation between the two and cited that previous researchers have found computer experience to be the precursor of computer self-efficacy beliefs. The following hypotheses were used in the present study:
H5. Computer experience will have a negative correlation with computer anxiety.
H6. Computer experience will have a negative correlation with computer anger.
Section snippets
Participants
Participants in present study include 242 students enrolled in various courses at a mid-sized, four-year public university (Mathematics course, n = 43; Health and Community Service courses, n = 101; Computer Science course, n = 30; Psychology courses, n = 68). The mean age of the participants was 22 years old (SD = 5.17). Females represented 57% (n = 138) of the sample and males represented 43.0% (n = 104) of the sample. All participants had previous computer use. Most participants were Caucasian (74.0%, n =
Results
Table 1 displays the means, standard deviations, and correlations between the independent variables and the dependent variables. Word processing and internet browser familiarity were the two domains which participants rated themselves as most experienced. The domain with the lowest reported experience was graphical editing software. All of the independent variables were negatively correlated with both computer-anxiety and computer-anger. Word processing was not significantly correlated with
Discussion
The purpose of this study was to discover which factors are related to a computer user becoming anxious or angry. This was not an experiment, and therefore it does not infer causation. Computer-use, computer-experience, and computer self-efficacy were all negatively correlated with both computer-anxiety and computer-anger. Thus, all six hypotheses were supported. However, they were supported to varying degrees. Computer self-efficacy clearly had the most significant impact on both computer-anger
Alleviating anxiety and anger symptoms
As previously mentioned, systematic desensitization (SD) would potentially reduce both computer anxiety and anger symptoms. The theory is based on the premise that an individual cannot be both fearful and calm at the same time. So, an individual is taught to remain calm while being presented with “the stressful stimulus.” Bandura (1986) also supports the claim that SD would be a useful technique in alleviating anxiety symptoms. Already anxious people probably need the assistance of a
Acknowledgements
This study was funded by a Research and Creativity grant from California State University, Chico. I thank Linda Kline, Department of Psychology, California State University, Chico for her invaluable support and guidance. I also thank Orlando Madrigal, Department of Computer Science, California State University, Chico for his encouragement.
References (22)
Computer anxiety: Relationship with computer experience and prevalence
Computers in Human Behavior
(2001)The relationship of instrumental and expressive traits with computer anxiety
Personality and Individual Differences
(2001)- et al.
Computer anxiety and its correlates: A meta-analysis
Computers in Human Behavior
(1999) The influence of specific computer experiences on computer self-efficacy beliefs
Computers in Human Behavior
(2003)- et al.
The effects of computer anxiety, state anxiety, and computer experience on users’ performance of computer based tasks
Personality and Individual Differences
(1997) - et al.
The development and beta-test of a computer-therapy program for anxiety and depression: Hurdles and lessons
Computers in Human Behavior
(2003) - et al.
Frustrating the user on purpose: A step towards building an affective computer
Interacting with Computers
(2002) Self-efficacy: Toward a unifying theory of behavioral change
Psychological Review
(1977)Social foundations of thought and action
(1986)- et al.
Course structure effects on student’s computer anxiety
Journal of Educational Computing Research
(1995)