Teresa Torres

Home

Childhood Factors That Encourage
Computer Usage in Adult Life

Teresa Torres
Stanford University
May 1999

Table of Contents

Introduction

Review of the Literature

Important Definitions

The Four Areas Of This Study

Methods

Results

Discussion

Appendix

Introduction

Computers are becoming more pervasive every day. They influence every aspect of our lives from the way we work to the way we play. The age at which children are being introduced to computers decreases every year. Schools across the nation are now computer-ready and many have direct access to the rapidly expanding Internet. The number of people with home computers is also growing daily. This rapid increase in computing raises a number of questions concerning equity. If computers continue to play a significant role in our daily lives, everyone will need to be computer competent. Can schools make up for the socioeconomic differences in exposure to computers at home? Does race and ethnicity have an effect on who uses computers? What can schools do to encourage children to become computer users? What roles do friends, family and role models play in encouraging computer competence? Do boys and girls approach computers differently? Are there inherent gender differences in computer usage? Dozens of questions could be enumerated. The question is how do we answer these questions.

Many studies have attempted to get a handle on these and related questions. They explore why girls and women, people from lower socioeconomic classes or ethnic minorities do not use computers. But these studies all suffer from the same underlying assumptions. Embedded within their approach is the assumption that their subjects do not use computers at all. Clearly, this is not the case. All sorts of people, including girls and women, people from lower socioeconomic classes and ethnic minorities, use computers. It may be the case that these groups use computers less, on average, than some other group but statistically this is very different from not using computers at all.

Recently, more and more studies are focusing on the gender gap in computing. Due in part to reports released by the American Association of University Women and the recent growth in the software industry for girls, computer usage by women and girls has come into the spotlight. But, again, most of these studies ask why girls do not use computers. Few look at the women and girls who are using computers and examine why they choose to use computers when so many of their peers do not.

It is a legitimate endeavor to explore why minority groups may use computers less, on average, then other groups, but it is a daunting task to get meaning out of such studies. There are endless reasons why someone may not be a computer user. It would be difficult to define a set of characteristics or experiences that deterred a majority of non-computer users from using computers. It seems intuitively more likely that we could define a set of characteristics and experiences that encouraged computer users to use computers. This is a much smaller subset of the population. Additionally, by focusing on all computer users, we avoid the risk of making broad generalizations about any specific set of people.

This study attempts to analyze questions of inequity and in some cases seek out answers by focusing directly on computer users with an emphasis on analyzing perceived gender differences. It is based on the underlying assumption that some people (men and women) have interests in computers and often pursue computer-related careers. The purpose of this study then, based upon this assumption, is to seek out the motivating factors in these peoples lives. Instead of asking Why dont women use computers?, this study asks, What encourages people to use computers? It attempts to narrow in on the social factors that help shape peoples conception of computers and computer users.

It investigates factors in subjects lives beginning with early childhood extending through the college years. The study focuses on four main areas that may have encouraged people to become computer users: social status, family influences, friends and personality and introduction and exposure to computers. These main areas will be analyzed in turn, but to do so we need to first review the literature on similar studies and draw upon some important definitions.

Review of the Literature

As mentioned above, most of the existing literature has focused on why women and girls do not use computers. Although this study takes a different approach, it draws upon the knowledge gained in previous studies. The following is a review and discussion of the main factors which have been shown to discourage women and girls and in some cases all people from using computers.

Experience and Exposure

Intuition would lead us to believe that exposure to technology would reduce barriers to becoming interested in technology. Ayerman and Reed (1996) found that not enough exposure and experience with computers led to computer anxiety. As will be discussed later, computer anxiety has been found to be one of the most significant deterrents to computer interest. Brosnan (1998), however, found that exposure may actually increase computer anxiety. He argues that exposure to computers for people with computer anxiety deters them from the task they are working on and does little to reduce the anxiety. Hawkins (1985) may shed some light on the results found in these studies. She argues that girls are not exposed to computers in supportive environments. It may be that only exposure to computers in supportive environments reduces computer anxiety. If this is the case, it appears that the source of the anxiety is misnamed. It doesnt seem to be the computer that is causing the anxiety, but the environment in which computers are placed.

Similarly, Lips and Temple (1990) found that computer experience was the most significant predictor of computer enjoyment for females, but was not found to be a major predictor in males. The reason for this difference is not clear. Lips and Temple (1990) speculate that women benefit more from experience because they have less computer confidence. But they also suggest that women may have a more difficult time in computer courses. Women are likely to be surrounded by male peers who do not think they are competent and they often have to overcome larger obstacles to even enroll in computer courses. The exact role that computer courses play seems to be unclear. Despite this confusion, the number of computer courses taken was the criterion Lips and Temple (1990) used to assess computer experience. Although they touched upon it by speculating that women have less computer confidence, it seems as if Lips and Temple failed to analyze other factors that may contribute to why women benefit more from experience.

The age at which a person is first exposed to computers also seems to be significant. Clements (1985) argues that early and frequent exposure to computers encourages usage. Brosnan (1998) agrees. He argues that the time of the first computer experience is critical to reducing computer anxiety later in life. This is a significant factor. In a number of studies it has been shown that boys are introduced to computers, on average, at an earlier age than girls (Brosnan, 1995). Boys are also more likely to have a home computer (Miura and Hess, 1983; Nelson and Cooper 1989). Enrollment in summer computer camps favors boys and the gap widens as the difficulty level and the cost go up (Miura and Hess, 1983). All of these factors encourage boys interest in computers and do nothing to encourage girls.

Computer Anxiety

Computer anxiety is an ambiguous term that is used by a number of studies to describe both a cause and an effect of the computer gender gap. Most of these studies draw upon the LIKERT scale, a system that asks the subject to rate a number of phrases on a sliding scale from strongly disagree to strongly agree, to assess attitudes and perceptions about computing. These studies associate negative attitudes with computer anxiety. Instead of grouping perceptions and attitudes into one ambiguous group called anxiety, each of these factors will be looked at independently. But first, the following is a discussion of the results from a number of these studies.

Brosnan (1998) spends a large portion of his book, Technophobia: The Psychological Impact of Information Technology, discussing both the causes and the effects of computer anxiety. He argues that computer anxiety will exist as long as computers are still visible in everyday life. He also states that people with computer anxiety do not fear the computer task, they just fear the computer. This is a critical point. Many people conclude that computer anxiety is the result of the fear or dislike for a specific kind of task. This is less often the case. Brosnan has found that it is the machinery itself that is instilling the fear. He argues that computer anxiety can not only result in avoidance of computers but it can also lead to impoverished performance on various tasks. He explains the causes of computer anxiety as being perceived loss of control, fear of negative evaluation, and unfamiliarity with the machine language. Some of these factors will be discussed later He also cites a number of correlates of computer anxiety including gender, computer experience, and academic achievement. But he refers to a number of studies that provide conflicting data.

Shaw (1995) makes a general report based upon her own research and observations that women and girls are generically more anxious than men and boys (48). She argues that this is the reason why women underachieve, fail to think of themselves as capable or apply for senior positions, [and] limit their aspirations.(48).

Morse and Daiute (1992) are critical of the results that come out of these studies, arguing that LIKERT scales are not an appropriate means to measure anxiety along gender lines. They (1992) claim that womens responses change depending upon the context of the given phrase. In their own study, which drew upon open-ended questions, Morse and Daiute (1992) found that girls do not have as much computer anxiety as many other studies have shown. They argue that a good portion of the computer gender gap is a result of the testing methods.

Perception of Computers

A number of studies report that girls do not view computers as appropriate for their own use. Instead the results show that computers are thought of as primarily in the male domain (Crook, 1994; Hughes, Brackenridge and MacLeod, 1987; Swoope and Johnson, 1985). Singh (1992) reports boys are more comfortable with computers than girls. Giacquinta, Bauer and E-Levin (1993) found that boys and girls have different perceptions of the use of computers: Boys see computers as recreational, a toy to be played with, while girls see computers as utilitarian, a tool to a get a job done. But all of these results vary depending upon the age of the subjects.

Nelson and Cooper (1989) surveyed 5th graders from three public schools. They revealed that enthusiasm for computers was equal for boys and girls. But girls saw computers as equally important for girls and boys, while boys thought that computers were strictly for boys. Both boys and girls, however, thought video games were designed for male use. The relationship between video games and computer usage is discussed later. Nelson and Cooper (1989) did find an interesting link between perception and attitude toward computing. They found that girls, who thought that computers were specifically designed for them and not necessarily for boys, had more positive attitudes toward computers and felt more competent when using computers than girls who thought that computers were more appropriate for boys.

In a study conducted by Swoope and Johnson (1985), 1st, 3rd, 5th, 7th, 9th and 11th graders were surveyed. Their first study asked boys and girls to rate whether computers were masculine or feminine. Before 7th grade, all students said computers were both masculine and feminine. By 7th grade, however, both girls and boys rated computers as predominantly masculine. Their second study, which relied upon LIKERT scales, produced much different results. All of the 1st graders had positive attitudes toward computers. But after 1st grade, far more boys than girls thought boys had a greater interest in computing. This difference in results may possibly be attributed to the findings in the Morse and Daiute (1992) study that criticized the use of the LIKERT scale in such studies.

These last two studies suggest that boys and girls learn at some age that computers are more appropriate for boys. It also appears as if boys learn this earlier than girls. But there is little research that examines where and when girls and boys pick up this perception. It is likely, however, that a combination of factors, including schooling, adolescence, peer and adult influences, and stereotypes, contribute to the formation of such perceptions.

Hughes, Brackenridge and MacLeod (1987) interviewed the same elementary school children over a 16 month time span. During this time, the children participated in a computer curriculum within their school day. At the beginning of the study, when asked Who likes computers more? 42% said boys, 10% said girls and 47% said girls and boys liked computers the same. Similarly, when asked, Who would be better at computers?, they received the exact same results. But by the end of the program, these statistics changed dramatically. After 16 months of using computers at school, 14% answered boys, 4% answered girls and 83% answered the same when asked, Who likes computers more? Again, when asked, Who would be better at computers?, 17% said boys, 2% said girls and 81% said both boys and girls. In this case, extensive exposure to computers helped change childrens perception of computers.

Computers, in schools and in the home, are often associated with math and science (Hawkins, 1985). It is possible that observed differences in perception are related to the gender gap in math and science. In fact, Hawkins argues that the solutions to the gender gap in math and science are analogous to the solutions for the computer gender gap (1985). Other researchers, however, are quick to point out the differences. Brosnan (1998) explains that the number of women going into engineering fields is on the rise, but the number of women majoring in computer science is dropping (Brosnan, 1998; Crook, 1994). Brosnan (1998) speculates, there must be something about computer science that keeps women away.

Perception of Self

A number of studies have claimed that boys and girls perceive achievement differently. They argue that boys perceive success as a result of their own ability while girls attribute success to luck. Boys attribute failure to situational factors while girls credit failure to their own lack of ability (Mount Holyoke College; Hawkins, 1983). Nelson and Cooper (1989), however, speculated that these differences would exist, but did not find them in their own study. Hawkins (1983) speculates that this difference in perception may contribute to perceived differences in math ability. She argues that math is a domain in which wrong answers are more likely to occur than in other domains. Boys may think they are better at math than they actually are because they are attributing their failure to situational factors. Meanwhile girls may be better at math than they think they are because they are attributing their success to mere luck instead of ability. This speculation is supported by a study conducted by Swanson . She found in a survey conducted in 1990, that boys and girls performed roughly equal on a math test. But 22% of boys and only 14% of girls agreed to the statement I am very good at mathematics. Despite equal performance, boys perceived themselves as better math students.

These differences do not just exist in early childhood. They persist throughout life. Men and women who majored in math and sciences, including computer science, were influenced by their ability (or inability) to perceive themselves as a scientist or engineer, and by whether or not they felt they could contribute to their field. These influences were at work when deciding whether or not to pursue an advanced degree in their subject area. Significant ethnic and gender differences were found in both of these areas (Grandy, 1994). Similarly, Sturm and Moroh (1995) looked at 5 years of computer science students transcripts and found that women were doing just as well as men throughout the computer science curriculum. They also found that women were outperforming men in all levels of math. A survey they conducted, however, showed that most students think men outperform women in calculus (Sturm and Moroh, 1995). This study shows a stark contrast between how women are actually doing compared to how they are perceived as doing by both themselves and their male peers.

The relation between math and computer science is not clear. But it is safe to speculate that if a difference in perception of achievement does exist and it affects perception of math achievement it is likely that it would also affect perception of computer achievement. The results, however, have not been all bad. Grandy found that female students in computer science did not rate any of their skills below those of their male peers (Grandy 1994). But this begs the question, did the men rate their skills above their female peers? It is not just female perception of their own ability that is likely to contribute to their performance but also the perceptions of them held by their peers, including their male peers.

Expectations and Desires

It appears that men and women want and value different things in life. Grandy, in his survey conducted in 1994, found a number of different preferences among men and women for job activities, definitions of success and social interactions. He found gender differences in salary expectations, importance of making a contribution to society and job activities. His survey revealed that men expect to make more money, on average, than women and rank salary as more important. He also found gender differences in the job environment.

Grandy found that women had stronger preferences for working with people over things whereas men had stronger preferences for working with math over words. Preferences for a competitive environment versus a cooperative environment were the same in both males and females. Males, however, valued technical challenge more than women. Women, on the other hand, valued contribution to society, pleasant coworkers and prestige and respect more than men. This survey, however, surveyed math, science and engineering majors. Computer scientists showed the smallest differences across the board based on gender (Grandy, 1994).

This study raises the following questions: where do these differences stem from and how do they influence choices in regard to pursuing careers and interests? If these differences stem from external influences and pressures then it is more interesting to investigate those external influences and pressures than to merely conclude that these differences exist. The second question analyzes the consequences of these differences in wants and values and indirectly the consequences of the external influences and pressures. Answers, or at least additional information, about these questions would help us understand why, for example, some people choose to pursue computer related careers and others do not.

Parents Perception of Child and Encouragement

A number of studies have shown that parents play a critical role in developing a childs interest in computing. Parents influence their children through both their own actions and the amount of encouragement they give their children. During a three year study that observed childrens home use of computers, kids primarily used computers to play video games (Giacquinta, Bauer, and E-Levin, 1993). The study revealed a number of factors that kept children from moving beyond the video game domain into computing. They found that children had little to no encouragement from their parents. In most cases, the parents were computer illiterate and didnt have the know-how to encourage their kids to use the computer.

In most homes the mother and father played two very different roles in relation to the computer, often having a gender-roled effect on their childrens perspective of computers. The father was always the decision maker when it came to computers, meanwhile the mother was distant from the computer (Giacquinta, Bauer, and E-Levin, 1993). Sanders and Stone (1986) report parents often assume computers are more appropriate for boys. As a result, they are more likely to purchase a home computer for a son than for a daughter. Sanders (1993) found that fathers and brothers use computers more than any other family members.

Parents influence, however, extends far beyond just how they act around or think about computers. Gipson (1997) found that parents perception of their childs math ability influences their childs interest in computers. She also found that parents fears of both math and computers are often passed down to their children and affect the childs attitude toward computers (Gipson, 1997). Parental influence continues through the college years. Grandy (1994) reports that twice as many females as males, going into graduate programs in science or engineering fields, said their mother was in a technical, mechanical or scientific occupation. No such correlation was found between the fathers career and the students graduate school plans (Grandy, 1994).

Role Models

Role models play a significant role in the lives of children. It comes as no surprise that very few women are in a position to serve as role models to young girls who aspire to work in a computer-related career or wish to pursue an interest in computers. Many researchers have reported that girls need such role models and they just dont exist (Fuchs, 1986; Stalker, 1983). Sanders and Stone (1986) claim computer users are mostly male and this does little to encourage girls to use computers. In fact, they argue, it actively discourages girls from using computers. Sanders (1993) also blames television. She argues that girls only see men in computer-related roles on television. Additionally, Sanders blames adult women. She claims that every time a girl sees a hint of boredom, lack of interest, frustration or incomprehension on a female adults face at the mention of technology, girls are sent the message that technology is not for them (Sanders, 1993).

Video Games and Software

Many studies focus on the computer as a machine. Few look at how people are using computers. But it appears that a large portion of the problem may exist in how computers are being used. Fuchs (1986) claims that software is not gender-neutral. Software, from video games to major applications to simple utilities, is designed with men in mind (Sanders and Stone, 1986). In a study that surveyed 157 middle school children, subjects were asked to categorize 75 software titles. A significantly greater number of titles were rated predominantly for boys (Miura and Hess, 1983).

Computer games especially, reinforce the notion that computers are a male domain (Stalker, 1983). Nelson and Cooper (1989) found that both boys and girls see video games as being for boys, although girls do see computers as appropriate for both sexes. Boys stronger interest in video games may lead to a stronger interest in computers overall (Nelson and Cooper, 1989). Nelson and Cooper also found that video game usage by boys directly correlated to computer usage, but no such relation existed for girls.

Littleton and Others (1993) conducted an experiment which took a game, King & Crown, in which all characters are male and is based on a masculine stereotyped quest and created a similar game, Honeybears, which has gender-neutral characters and a fairy-tale scenario. Both King & Crown and Honeybears contained similar tasks, structures and cognitive demands. Littleton and Others, observed children playing both games and surveyed them afterwards on enjoyment of play. All gender differences that appeared with King & Crown were ameliorated with Honeybears (Littleton and Others, 1993).

Stereotypes, Friends and Peer Pressure

Brosnan (1998) argues that the most significant correlates of computer anxiety are gender and expectations in relation to gender. Other researchers agree. Gipson (1997) claims that societys stereotypes, both formal and informal, affect attitudes about computing. Young children, according to Davidson and Davidson (1994), need to categorize both people and objects. To young children, differences stand out while similarities are often overlooked. From an early age they begin to categorize the people and objects around them by gender. Huston (1985) as quoted in Brosnan (1998) claims sex-role stereotypes begin to develop at age two and are strongly established by age seven.

This categorization is reinforced in books, on TV (Stalker, 1983) and by parents and friends. Mischel (1970) as quoted in Brosnan (1998), argues, observational learning and reinforcement by others are the essential elements in developing a childs sex-type. Swanson agrees. She argues that social environments have the greatest impact on personality, play preferences, and learning abilities. Swanson claims, we teach our children to act like boys and girls. We do so by doing such things as telling boys to explore and girls to be careful (Swanson).

With the onset of adolescence the confirmation of these categories becomes a central theme in the lives of children. Angell (1991) claims that girls (but should also include boys) quickly learn that some behaviors and attitudes are more important for one sex than the other. Also during this time, parents lose a piece of their influence as peers influence grows exponentially (Davidson and Davidson, 1994). Sanders and Stone (1986) agree. They explain kids look to their peers to help them figure out who they are. Most kids feel that they are limited to characteristics that are associated with their own sex. For girls, this just about rules out computers since most kids think computers are exclusively for boys. As Sanders (1993) explains, the computer gender gap is reinforced when the popular girls are not in the computer club.

Nelson and Cooper (1989) explain the effect this categorization has on children. They found that sex-typed individuals are likely to avoid public cross-sex behavior (Nelson and Cooper, 1989). For girls, this means computers are off limits. Barrie Thorne in Gender Play (1997) also discusses gender categorization and cross-sex behavior. She argues that in the earlier years both boys and girls occasionally participate in cross-sex behavior but not without facing the criticism of their peers. In the later years, more girls participate in cross-sex behavior than boys. But most only participate in the periphery, mostly as observers. It is difficult to develop the necessary skills to become an active computer user as an observer. Thorne explains, only the skilled girls successfully penetrate cross-sex behavior boundaries.

The School System

The school system plays a large role in influencing children's interests and attitudes toward computers. It can both contribute to the problem and alleviate it. Schools affect how children perceive computers through placement of computers, scheduling of computer classes, teachers' attitudes and expectations and student interactions and expectations.

Computers are often found solely in math and science classes (Stalker, 1983). While it is good that computers are being used in the classroom, such placement introduces an array of problems. Math and science classes already suffer from gender differences in how the classes are perceived (Stalker, 1983; Sanders and Stone, 1986). Introducing computers into these areas does two things. First it adds whatever stigma is attached to the computers to the math or science classroom and second it is possible that the existing gender gap in math and science is contributing to the gender gap in the world of computers. Sanders (1993) explains, schools also contribute to the problem by scheduling computer classes at the same time as music and art classes.

Teachers attitudes and expectations have lasting influences on their students. Teachers teach predominantly to the boys (Mount Holyoke College, Sadker and Sadker, 1994). Sanders (1993) adds, teachers ask only the boys technical questions and wait longer for boys to think of answers. Additionally, most teachers assume computers are more appropriate for boys (Sanders and Stone, 1986). Gipson (1997) explains, these expectations by teachers affect computer attitudes.

The school experience as a whole also has an impact. Crook (1994) speculates we may say that gender-based attitude differences are not convincingly present at the start of schooling; they must somehow be cultivated within the early school years (Crook, 1994). Scheingold, as quoted in Brosnan (1998) also argues that elementary schools deny girls credibility as computer users. Connell, Ashenden, Kessler and Dowsett (1982) implicate the school system for far more than just effects on computer usage. They argue that school plays a critical role in the production of masculinity and femininity.

Possible Sex Differences: Cognitive and Social

In a society riddled with gender stereotypes, as we saw above, the area of computer usage is not exempt from the categorization of such stereotypes. But is there truth to any of these stereotypes and if so, to what extent? A number of researchers have attempted to assess cognitive differences between the sexes and many more are quick to name the social differences between men and women.

Brosnan (1998) reports on a number of studies that investigate cognitive differences among people. He quotes Caplan et al (1985) who found that differences in spatial perception tend to be small and inconsistent across tests. Griffiths (1985), also quoted in Brosnan, argues that none of the proposed biological differences have stood up to critical scrutiny. Most have been either trained away or shown to have never existed.

Witkin and Goodenough (1981), in Cognitive Styles: Essence and Origins: Field Dependence and Field Independence, make a cognitive distinction between field dependence and field independence. They conducted experiments in which a subject had to determine when a moving rod was perpendicular to the ground. There were two cases: one in which the subject relied upon visual input to determine when the rod was upright and the other the person relied upon the direction of gravity. Some subjects only relied upon one of the two methods, but most drew from both to make their conclusion. All subjects were consistent across trials. Those who relied more on visual input were field dependent meanwhile those who relied more upon the direction of gravity were field independent.

Field dependent and field independent people often perform differently on a variety of tasks. Field independent people have no problem picking an image out of a complex background, but field dependent people have difficulty with this task. According to Witkin and Goodenough, field independent people may be better at perceptual constancy, speed of closure, functional fixity, conservation, representation of the horizontal coordinate, perspectivism and picking out lexical ambiguities in sentences. Field independent people function more autonomously, meanwhile field dependent people tend to be more interpersonally oriented. A number of studies attribute field independence to men and field dependence to women, but results are not consistent across studies. This is especially difficult to claim, since most people rely upon both methods to some degree.

Maccoby and Jacklin (1974), in Psychology of Sex Differences, discuss a number of possible cognitive sex differences. Most of their studies were on infants to avoid socialization factors. They found no sex differences in any of the five senses, in conditioning, paired-associates learning, discrimination learning, delay of reinforcement, partial reinforcement, incidental learning, learned low-amplitude responding, probability learning, learning though imitation, memory and social memory.

In an attempt to tackle many of the popular stereotypes, Maccoby and Jacklin found no evidence that girls are more interested in social stimulus than boys. Some studies found sex differences in verbal abilities, favoring women, but most studies found none. In all studies that did find differences, inconsistencies existed across age. They found no sex differences in quantitative abilities in the early years. Differences favoring boys began to appear between ages 9-13. For spatial abilities, some studies found girls had the advantage between ages 3-5, while boys had the advantage between ages 6-8. But most studies found no differences in any ages. In no area did Maccoby and Jacklin find significant sex differences that were present in infants and were consistent across all tests and studies.

The Effects of Research

After conducting a study, analyzing a number of the studies previously mentioned, Morse and Daiute (1992) warn "...that building expectations about differences can actually create differences." (2). They report that studies are not consistent in defining common terms, having found 15 definitions of computer attitudes, 8 definitions of computer aptitude and 7 definitions of computer usage in among 55 studies. They also argue that many studies put too much emphasis on differences. Morse and Daiute (1992) argue that often times the similarities are greater than the differences, but it is the differences that the researchers are looking for. Campbell in 1988, as quoted in Morse and Daiute (1992) claims "gender research tends to focus on the deficit model reporting who's better and whos worse, often ignoring or overlooking how the genders might be similar." (3). Morse and Daiute (1992) also criticize other studies for their lack of "statistical rigor. They argue that LIKERT scales, which most studies rely upon, are inaccurate opinions and are dependent upon context which many of these studies lack. They also argue that most studies focus on the computer not on how the computer is used. In their own study, Morse and Daiute found that girls are more enthusiastic and enjoy computers more than others studies have shown. They also found that girls showed a lot less computer anxiety when given the opportunity to answer open ended questions within a given context.

Discussion of the Literature

These studies have provided us with a wealth of information about factors that discourage people, and in many cases women, from becoming comfortable computer users. Parents and teachers lack of encouragement, a scarcity of appropriate role models, peer influences, little exposure to computers and male dominated software have all been exposed as barriers to computer usage for people, especially for women and girls. This study hopes to use these known barriers and examine the flip-side. To what extent do people benefit from encouragement, role models, friends, exposure to computers and video games? If we introduce these elements into a persons life will they be more likely to be a computer user?

We can also draw upon the existing research to define who is comfortable using computers. We have learned that computer anxiety may exist even in those who do use computers. We know that people perceive computers differently, some people have difficulty seeing themselves as computer users, and others want and value different elements in the work place. In defining who is and who is not a computer user, we need to account for computer anxiety, career interests and peoples perception of computers.

Important Definitions

Computers

The term computer often encompasses anything that comes equipped with a silicon chip, from you desktop computer to your newly automated car. In this study, however, the term computer is used only to refer to personal computers including desktops, laptops, personal data assistants and video game console machines. Older models of computers are also intended to be included in this term (ex.. Atari, Commodore, Apple+, etc.).

Computer Users

What qualifies a person as a computer user? In this study there are five main criteria that determine whether or not a person is a computer user: enjoyment of computer usage, being comfortable working with computers, using computers on a daily basis, use of computers on a daily basis even if it was not required, and proficiency in a computer language. This study attempts to distinguish between people who use computers because they enjoy using computers and people who use computers because it is required. It also distinguishes between people who are merely computer consumers and those who are computer innovators.

This first distinction is pretty clear. Many people use computers on a daily basis but would not continue to use computers on a daily basis if it were not required. These people may not dislike using computers but they see no need to continue using computers outside of specific computer-related tasks. Others use computers on a daily basis even though it is not required of them by their job or would continue to use computers on a daily basis even if it was not required. This study aims to pick out those who use computers because of some intrinsic value of the computer not because of the value of the context in which the computer is being used. This distinction also helps to factor out those who may use computers regularly, but still have computer anxiety. Odds are, these people do not enjoy using computers, are not comfortable using computers, or would not continue to use a computer on a daily basis if it were not required.

The distinction between computer consumers and computer innovators, however, needs some clarification. Many people use computers on a daily basis at work or even at home. Some users, for example, word process all day, check e-mail and occasionally surf the Internet. These people fall into the category of computer consumers. Other people use computers to create interactive web sites, write shareware or commercial software and extend the use of their computer beyond what the provided software can do. These people fall into the category of computer innovators. This study attempts to distinguish between those who use computers as a tool, but do not know enough about computers to use them beyond a well-defined purpose or would not know how to analyze a problem if one were to occur, and those who are confident enough to stretch their computers to the limit and would know how to debug a problem when one arises. A computer user is defined as fulfilling all five criteria. In other words, a computer user:

  • enjoys using computers.
  • is comfortable using computers.
  • uses computers on a daily basis.
  • would continue to use computers on a daily basis even it were not required.
  • is proficient in a programming language.

Programming Languages

As defined above, a computer user is proficient in a programming language. In order to avoid ambiguity, a programming language will be defined as any language, either interpreted or compiled, that can be used to create executables. This includes BASIC, C, C++, Pascal, and Java but is not limited to this set. It does not, however, include HTML. The rationale behind this decision is that in all of the previously mentioned languages, the user is telling the computer what to do. In order to do this, the user must overcome a fear of breaking the computer. Additionally, the user is instructing the computer to do something, instead of the computer instructing the users. HTML does not stretch the users conception of a computer. Task-wise, it is no different than learning a word processing program. Additionally, many people confuse knowledge of HTML with knowledge of an HTML editor. To avoid categorizing people who are proficient in Claris Home Page or Netscape Composer as proficient in a programming language, HTML is not considered a programming language in this study.

The Four Areas Of This Study

As mentioned before, this study focuses on four main areas: social status, family influences, friends and personality and the subjects introduction and exposure to computers. Survey based research, described below, was conducted in an attempt to analye the many elements in these four areas. Each of these areas will be analyzed in turn.

Social Status

Ethnicity, age, gender and income are all likely to affect exposure to and interest in computers. Ethnicity is bound to play a complex role in determining potential computer users which undoubtedly will be beyond the scope of this study. But it is safe to predict that ethnic minorities will be at a disadvantage in both their access to technology and the encouragement they receive in pursuing interests in computers. This is likely tied to other aspects of our society that are influenced by race and ethnicity such as education, employment and opportunity. Additionally, it is likely that race and ethnicity will be intertwined with many of the other social status elements addressed below.

Age is also likely to effect who is bound to be a computer user. Those who grew up with computers are more likely to be computer users than those who were introduced to computers later in life. But in time, this factor will probably be less significant. This study, however, only looks at college-aged students. College is a period in which people are deciding what they want to do in life. Pre-college subjects may not be able to predict as well whether or not they will be computer users later in life. Post-college subjects are less likely to remember their childhood experiences. College students are also being introduced to a number of new things, possibly including computers. There will likely be little variance by age, however, in this study, although younger college students may have been introduced to computers at an earlier age.

As a number of studies have shown, gender plays a significant role in determining who becomes a computer user. As mentioned above, boys are expected and encouraged to use computers while girls are often denied the opportunity to explore their computer interests. As a result, more men pursue their interests in computing and are represented by larger numbers in computer science programs and in the computer industry. It is likely that this study will find more men claiming to be computer users than women.

Income is also likely to play a significant role in determining who becomes a computer user. People with lower incomes are less likely to have access to computers and are less likely to have computer literate parents. People from lower income backgrounds will most likely be introduced to computers much later in life and probably given little to no encouragement to pursue computer interests. They are also less likely to own a home computer.

Family Influences

Parents play a critical role in forming their childs computer experience. Children are likely to be influenced by their parents beliefs and attitudes toward computers. Well-educated parents, especially those in computer-related careers, are more likely to encourage their children to become computer users. Similarly, children of computer-literate parents, specifically those who use computers in the home, are more likely to become computer users. Children, especially daughters, with mothers who use computers at home may also be more likely than children with just fathers who use computers at home to become computer users themselves. Girls who grow up in homes in which only their father uses the computer may not receive the message that computers are also appropriate for women and girls.

Siblings are also likely to have a significant effect on computer interests. Children with older siblings are bound to be influenced by their older siblings interests and hobbies. People who grew up with older siblings who used computers in the home may be more likely to be computer users. There may be a similar correlation with younger siblings. But it is likely that it will not be as strong. Older siblings are more likely to act as role models to younger siblings. Of course, this is contingent upon the amount of time spent with older siblings. Older brothers may have a larger impact on computer interests than older sisters simply because older brothers are more likely to be computer users themselves.

Friends and Personality

Role models are likely to play a significant role in encouraging a child to become a computer user. Children with accessible role models who use computers and encourage the child to use computers, especially girls with women role models, may be more likely to become a computer user than those who lack such influences. An accessible role model is meant to define someone who interacts with the child in day to day life. It could be anyone from a family member to a friend. Role models may help a person overcome barriers that have been built up in relation to computing.

Children are also bound to be influenced by their peers. Children, as was discussed before, often turn to their peers to categorize appropriate behavior and activities. People who have friends who use computers, especially during the critical period beginning in elementary school and continuing through middle school, may be more likely to become computer users. This is especially likely to be the case if same-sex friends are computer users. Girls may have many male friends who use the computer but still not see computers as appropriate for girls and thus not use them. Another critical period in which friends may have a strong influence is during the college years. This is a time in which people are being introduced to an array of new things including computers. It may be that people who have friends in college who use computers may be more likely to become computer users themselves.

Likewise, women who have predominantly male friends may be more likely to become computer users. This is likely to be a direct result of the fact that it is more likely that their friends will be computer users. But women who considered themselves tomboys growing up may be even more likely to become computer users than those who simply played with boys. When defining oneself as a tomboy, a person has already become accustomed to participating in cross-gender activities. It is likely that they will have less barriers to pursuing interest in computing because they already expect themselves and may be expected by others to pursue cross-gender interests.

Similarly, the number of friends a person has may be related to computer usage. But it will likely depend upon how much time a person spends with which friends and the habits of those friends in relation to computers. But it is probably safe to speculate that shy boys will be more likely to become computer users than outgoing boys. Shy boys may find security in the solitude of a computer. Shy girls, on the other hand, are probably less likely to become computer users than more confident girls, because they have to overcome the gender-typed barrier surrounding computers.

Introduction and Exposure to Computers

The subjects introduction and exposure to computers are also bound to have a significant effect on whether or not that person becomes a computer user later in life. People who are introduced to computers early in life and grow up with computers are more likely to be computer users than those who are introduced to computers later in life. People who grow up with computers have less fears to overcome while people who are introduced to computers later in life may have more challenges in learning how to use a computer. The earlier a person is introduced to computing, the less likely he or she will be influenced by societys perceptions of computing and societys perception of computer users.

Similarly, people who are introduced to computers in conjunction with another hobby are more likely to become computer users. If people are introduced to computers in light of another activity, they see an immediate purpose for the computer instead of having to find some intrinsic value in computers and as a result may be more likely to continue to use computers later in life. For children, especially boys, being introduced to computers via video games is often an excellent way to pique their interest.

Methods

In an attempt to address the hypotheses stated in the previous section, a quantitative survey of undergraduate students at Stanford University was conducted. 135 students were asked to respond to a survey which asked them questions about each of the details described above. The subjects were sophomores, juniors or seniors currently enrolled at Stanford University. Freshman were not surveyed for two reasons: many freshmen have not yet chosen a career path and often times freshmen discover and often pursue computer interests later in their college career. Surveys were distributed by residence in order to get a random sampling of students from various academic departments. All surveys were anonymous.

Stanford students were used as the subject pool for the following reasons. Although most students on campus use computers on a daily basis, it appears as if there is a strong split between computer users (as defined above) and non-computer users. Students at Stanford University categorize themselves as either technical or humanities-based. This is often a self-categorization and more often than not affects a persons academic interests. It is also likely to be a strong indicator of whether or not a person is or will become a computer user. Despite the ubiquity of computer usage on campus, not everyone becomes a computer user.

Stanford students, as subjects, do present a set of disadvantages, however. Although the University strives to achieve diversity on campus, this is a difficult task to accomplish while maintaining academic excellence. Many variables, especially those related to academic performance, ambition, parents occupation and level of education, may not be an accurate representation of the population as a whole. For example, a student from a low income level, may be more likely to use computers than a non-student from the same income level for the very same reasons that student ended up at Stanford. But despite these differences, there is still a significant difference between computer users and non-computer users at Stanford that deserve investigation. Further discussion of the relation between the sample pool and the population as a whole will occur in the results section.

The questionnaire was designed based upon previous research reviewed above and a dozen one-on-one interviews with current Stanford students. All of the interviewees were classified as computer users as defined by the criteria defined above. The interviews were used to pick out possible factors that may have encouraged computer usage that have not been considered in previous studies.

Results

The questions on the questionnaire fell into three categories: yes/no questions (ex. Did your mother encourage you to use computers?), categorical questions (ex.Mothers Education) or the time specific questions (ex. Most of my friends were boys: Before Kindergarten, During Elementary School, During Middle School, During High School and During College). The yes/no and categorical questions were coded as such and were tested against the computer user variable via a test of independence (Chi Square). The time specific questions had a more complex analysis. For each question there were a total of six data points: one continuous variable that summed up the number of time periods in which the subject marked yes and five yes/no variables that correspond to each time period. The yes/no questions were analyzed via tests of independence (Chi Square) while the continuous variable was analyzed via a test of significance (Unpaired T-Test). Throughout the discussion of the results, the five yes/no variables will be discussed in relation to the time period they represent. The continuous variable is treated as an overall indicator of how often the subject answered the question affirmatively (ex. How much the subject programmed or for how long the subject had computers at home, etc).

Each variable is analyzed in relation to the entire sample population, the male population and the female population. Often times, variables that are significant or dependent in the total population only show up in one or neither of the gender populations. In cases where it is only significant in one of the populations, the data is interpreted as being more significant in that one population but not necessarily insignificant in the other. In instances in which the data is significant in the total population but neither the male or female populations, the variable is interpreted as significant, but no judgement is made whether or not it is more significant for either gender. All tests are assessed at a 95% confidence level.

Social Status

Ethnicity appears to have no direct effect on computer usage among Stanford students (Direct Factors, Appendix). A couple of factors may have led to this result. First of all, about 10 percent of the sample pool chose not to fill out the ethnicity question, limiting the sample population. Additionally, ethnic minorities at Stanford may not necessarily represent the larger population of ethnic minorities. As was speculated above, it is likely that more ethnic minorities at Stanford are computer users for the same reasons that they are at Stanford in the first place. The opportunities and factors that may have helped/encouraged them to come to Stanford are also likely to play a role in their computer usage. Also, ethnicity may not significantly influence computer usage on its own. Coupling income with socioeconomic class or cultural influences, however, may bring out the effects of ethnicity. Or these other factors may account for observed ethnic differences in prior studies.

It was also hypothesized that ethnic minorities may have had less access to technology and less encouragement to use computers from their parents and role models. The data, however, does not support the claim that ethnic minorities at Stanford have less access to technology. There was no significant difference between ethnic minorities and the rest of the sample pool in either owning home computers or using computers at home. (Indirect Factors, Appendix). The data does support the argument that ethnic minorities receive different amounts of encouragement from parents and role models, but not as was hypothesized. Ethnic minorities actually received more encouragement, on average, than the rest of the sample pool from male role models before kindergarten, during elementary school and during college and from female role models during elementary school. Women minorities, on average, receive more encouragement than male minorities from male role models before kindergarten and during elementary school . No differences in encouragement from parents, however, was evident (Indirect Factors, Appendix).

A number of explanations could be attributed to why ethnic minorities receive more encouragement in relation to computing than others. Ethnic minorities may be more likely than others to have role models, or receive more encouragement from role models, as a result of various mentoring programs or adults perception of what minority children need. Additonally, role models can be anyone including, but not limited to parents, older siblings, teachers and family friends. Ethnic minorites may receive more encouragement through school programs or from older siblings. Encouragement from role models may be more important to ethnic minorities and as a result they may be more likely to report having role models who encouraged them to use computers. Because the data is dependent upon self-reports it is difficult to assess whether or not minority children actually receive more encouragement.

It was hypothesized that computer users would be younger than non computer users. But this did not turn out to be the case. The mean age for computer users was significantly greater than the mean age of non-computer users. When the population was divided by gender, however, this age difference only persisted in the female population. No significant age difference was found between male computer users and male non computer users (Direct Factors, Appendix). It was also hypothesized that younger students would be introduced to computers at a younger age, but no significant correlation was found (Indirect Factors, Appendix).

Intuitively, one might argue that older students have had more time in college to be introduced to programming, which may explain why older students are more likely to be computer users. There is a correlation between older students and the amount of programming experience, although when split by gender, it is evident that this correlation is much stronger for male students and may not be significant in the female population (Indirect Factors, Appendix). This is not a good explanation of why older students are more likely to be computer users, since this age difference only appeared among female students.

Additionally, only part of the correlation between age and programming experience is explained by exposure to programming in college. Older students, on average, are more likely to have programmed during elementary school, middle school, high school and college. However, older men are more likely than older women to have programmed during elementary school, middle school and college (Indirect Factors, Appendix). Exposure to programming in college does not seem to fully explain the difference in age. Nor does it explain why older women are more likely to be computer users than younger women.

The difference in programming experience from elementary school through college among older and younger students may be a result of the use of computers in schools. It is possible and very likely that a few years difference, makes a significant difference in how computers are being used in the classroom. When older students were in grade school, it is likely that computers were only used for programming, whereas a few years later when younger students were in grade school, computer classes did not have to rely upon programming alone. During this short period, it is likely that the use of commercial applications in schools increased, ultimately leading to a decrease in programming in school. Additional data, however, would have to be collected to support this speculation.
Gender is shown to be a significant factor in determining computer usage among Stanford students. As was hypothesized, males clearly have the advantage over females in becoming a computer user (Direct Factors, Appendix). Gender differences also existed in a number of the other variables as will be discussed below.

Parents income alone does not appear to be a significant factor in determining computer usage (Direct Factors, Appendix). This is likely to be a result of similar factors described in relation to ethnicity. It is likely that low income students at Stanford, were given additional opportunities to become computer users that counterbalanced the disadvantage of coming from a low income home. Again, Stanford students from low income homes are not necessarily representative of the low income population as a whole.

It was also hypothesized that income may influence access to technology for both the subject and the subjects parents. There were few differences, however, in overall computer usage or ownership based upon income. The only people that may be at a disadvantage are those in the $0-25,000 range. But differences only existed in home computer ownership, not in computer usage and they do not appear to be at all significant in the female population. It is not clear why low income males appear to be at a disadvantage in owning computers, when low income females do not appear to suffer from the same disadvantage. Neither group seem to suffer from a lack of opportunities to use computers.

Parents computer usage also seemed to be unaffected by income. The only significant difference in either parents home computer usage between income brackets was that a significantly greater number of mothers in the $90-150,000 income range used computers during subjects college years (Indirect Factors, Appendix). Additionally, there were no significant differences in encouragement from either parents or role models between income brackets as was hypothesized.

The only income bracket that presented a significant difference in the age at which subjects were introduced to computer was the lowest bracket of $0-25,000. Subjects from this income bracket were introduced to computers significantly later in life than subjects from most other income brackets (Indirect Factors, Appendix). This may be related to the groups disadvantage in owning home computers. Children in low income home may be introduced to computers outside of the home, escalating the age of introduction.

Family Influences

It was hypothesized that parents have a stronger impact on their childrens interests in computers than the data actually supports. It was hypothesized that well-educated parents, especially those in computer related careers, would be more likely to encourage their children to use computers, ultimately leading to a higher rate of computer users. But the data does not support either parents education or parents occupations as affecting whether or not a subject is categorized as a computer user (Direct Factors, Appendix). Nor is there a significant relationship between parents education and occupation and the amount of encouragement parents give their children in relation to using computers (Indirect Factors, Appendix). Additionally, there is no significant difference between computer users and non-computer users in how much encouragement they received from either parent, contrary to findings and speculations in earlier studies (Direct Factors, Appendix).

These findings are not immediately surprising in light of the many studies that have shown that parents influence is not as strong as once was speculated. It is surprising, however, that there is no significant relationship between a parents education or occupation and the amount of encouragement they give their child to use computers. Parents occupations were categorized as either technical or non-technical. Technical fields included all computer industry fields and many engineering fields but not necessarily science fields. All other fields were considered non-technical. Despite the intuition that parents would be more likely to encourage their children to use computers if they were in a computer-related career themselves, it appears as if this is not the case. However, it is not safe to conclude that parents who work in technical fields do not encourage their children to use computers more than parents who do not work in technical fields. The number of subjects who had parents who worked in technical fields was too small (less than ten) to make a conclusive statement about the relationship between a parents occupation and the amount of encouragement a parent will give his or her child in relation to computing.

Time spent with either the mother or the father also does not seem to be as significant as was hypothesized. Before college, time spent with parents appears to have no direct effect on computer usage. But women who spend time with their fathers during college are more likely to be computer users whereas women who spend time with their mothers during college are less likely to be computer users. There is no similar relationship for male subjects and their parents. Subjects who have parents who use computers during college are also more likely to be computer users (Direct Factors, Appendix). It is not clear whether parents computer usage during the subjects college years encourages the subject to become a computer user or whether the subject being a computer user influences the subjects parents computer usage. It is clear, however, that parents computer usage does not significantly impact computer usage earlier in the subjects lives. Many researchers have argued that children benefit from seeing their parents use computers, specifically in the case of girls who grow up with mothers who use computers (Giacquinta, Bauer, and E-Levin, 1993; Sanders, 1993). But this study does not support such arguments. Male subjects who have mothers who use computers while the subject is in middle school are less likely to be computer users (Direct Factors, Appendix). This result will be discussed below in the context of female role models.

It was hypothesized that older siblings who used computers at home would have a significant impact on whether or not a subject would be a computer user. The data, however, does not support any such differences. Older siblings home computer usage and the amount of time spent with older siblings appears to have little, if any, impact on whether or not a subject becomes a computer user (Direct Factors, Appendix). The implications of this finding rely upon the relationship between subjects and their older siblings. It could be interpreted very differently depending upon whether or not the subject admired or enjoyed spending time with his or her older siblings. The survey in this study, unfortunately, did not ask for such data.

Younger siblings, however, do seem to impact who becomes a computer user. Oddly enough, the data supports the claim that subjects who have younger brothers are less likely to use computers, even though there is no such relationship with younger brothers home computer usage or the amount of time spent with younger brothers (Direct Factors, Appendix). It is difficult, however, to assess how significant this relationship is because it only exists in the total population. The relationship is not significant in either the male or female population alone.

Male subjects do seem to be significantly influenced by their younger sisters. Males who had younger sisters who used computers while the subject was in middle school and high school are more likely to be computer users than subjects who did not have younger sisters who used computers during these time periods (Direct Factors, Appendix). It is not clear if male subjects are motivated to become computer users because their younger sisters are using computers or if their younger sisters are using computers because their older brothers are computer users. But, as was mentioned above, there was no significant relationship between being a female computer user and having older brothers who used computers. Similarly, all subjects who spent time with their younger sisters during college were more likely to be computer users than those who either do not have younger sisters or did not spend time with their younger sisters during college (Direct Factors, Appendix). More context about family dynamics in individual situations would be necessary to make further judgements.

Friends and Personality

It was hypothesized that role models would have a significant impact in encouraging subjects to become computer users both in their direct encouragement and by serving as example computer users. Subjects who had male role models who used computers were more likely to be computer users than those who did not, but this relationship was much stronger for females than for males. Male role models who used computers influenced computer usage during all periods of life until college. Women, specifically, were significantly influenced by male role models who used computers during college, but not during high school. Women were also influenced by encouragement from male role models during elementary school, middle school and college. For the sample as a whole, however, encouragement from male role models was more influential during high school than in college (Direct Factors, Appendix).

The role that female role models play in motivating computer usage was overestimated by the hypotheses. It was hypothesized that female role models who used computers and/or encouraged subjects to use computers would play a significant role specifically in the lives of female computer users, more so than non-computer users. But no such relationship is evident in the data. No significant relationship exists between the presence of female role models in the total population or the male or female population. Female role models do appear to play an interesting role in the lives of middle school and high school boys. Although the data does not support a conclusive significant difference, it appears that middel school and high school boys who have female role models who use computers or encourage the subject to use computers are less likely to become computer users than male subjects who do not have female role models who encourage them to use computers. (Direct Factors, Appendix).

These findings are consistent with the roles that parents play in encouraging computer usage. Male role models are more influential in the lives of women, especially during college, whereas earlier we saw that fathers play a significant role in the lives of college women. Despite all the research arguing that women need female role models (Sanders and Stone, 1986; Fuchs, 1986; Stalker, 1983), male role models appear to have a stronger impact in the lives of women. Similarly, we saw that mothers who used computers during a male subjects middle school years, were likely to negatively influence the childs interest in becoming a computer user. It appears as if female role models may play a similar role in the lives of both middle school and high school boys. This is a difficult result to explain from the current data set. Pure speculation might lead one to attribute it to gender stereotypes. Middle school and occasionally high school, tend to be time periods in which children are categorizing activities as either gender appropriate behavior or not. Boys who see women using computers may be less likely to categorize computers as an appropriate activity for boys. But further research would be required to support this claim.

Friends appear to play a significant role in determining who will ultimately become a computer user as was hypothesized. Their role as influences, however, is not limited to just computer use. It was hypothesized that subjects who had friends who used computers, especially during elementary school, middle school and college would be more likely to be computer users. But this was rarely the case. The only time period in which friends using computers had such an impact was during middle school for male subjects. There were no significant differences between computer users and non-computer users during all other time periods (Direct Factors, Appendix).

Friends who played video games were a stronger influence on subjects than friends who used computers but still not as strong as was hypothesized. Subjects who had friends who played video games during the subjects high school and college years were more likely to be computer users than subjects who did not have friends who played video games during these times. College is a particularly influential time for women. Women who had friends who played video games in college are much more likely to be computer users than those who did not have such friends (Direct Factors, Appendix). Other time periods were likely not as important, since most young children play video games. There were no significant relationships between subjects who had more male friends who played video games or who had more female friends who played video games and subjects computer usage. Almost all subjects marked having more male friends who played video games, even during time periods in which they did not indicate that their friends played video games (Direct Factors, Appendix).

Subjects who claimed that most of their friends were boys during elementary school and middle school were more likely to be computer users than those who reported otherwise. Male subjects who reported most of their friends were men during college were less likely to be computer users whereas female subjects who reported most of their friends were men during college were more likely to be computer users. Similarly, subjects who reported most of their friends were girls during elementary school and middle school were less likely to be computer users. Although not conclusive, men who reported most of their friends were females during college appear to be more likely to be computer users (Direct Factors, Appendix).

There was no significant relationship between women who reported being tomboys and women who were categorized as computer users, contrary to what was hypothesized. Shyness, however, did prove to play a significant role in the lives of those who were categorized as computer users. It was hypothesized that shy men would be more likely to be computer users whereas shy women would be less likely to be computer users. But the data supports the claim that both shy men and shy women are more likely to be computers. For men, shyness is correlated with computer usage during all periods in life. Shyness is correlated with computer usage up through middle school for women (Direct Factors, Appendix).

Shyness also influences who a subjects friends will be. Shy girls in elementary school and middle school are less likely to report that most of their friends are male whereas shy boys are more likely to report that most of their friends are girls in middle school (Indirect Factors, Appendix). These findings complicate the factors that influence computer usage. It appears as if the same factors that may positively influence computer usage may also have some negative effects. It was reported earlier that subjects who had mostly male friends during elementary school were more likely to be computer users than those who did not. Shy girls are less likely to have male friends in elementary school and middle school but they are more likely than girls who are not shy to be computer users.

Introduction and Exposure to Computers

It was hypothesized that the earlier a subject was introduced to computers the more likely that subject would be a computer user. The data, consistent with other studies, supports this claim, but the relationship is much stronger for males, and may not be significant for females (Direct Factors, Appendix). Other studies found that an early introduction reduced anxiety and encouraged usage (Clements, 1985; Brosnan, 1995). But this may only hold for male subjects. This study did not find that men were introduced to computers earlier than women, contrary to what other studies have found. Yet, men seem to benefit more from the same early introduction women are receiving.

It was also hypothesized that ownership and use of computers in the home would influence a subjects likelihood of becoming a computer user. Overall, home computer usage and ownership did contribute to becoming a computer user, particularly during elementary school and high school. This relationship was especially influential for males during these time periods and in college. The impact these factors had on female subjects, however, may not be significant (Direct Factors, Appendix). Again, men seem to be benefiting more from a similar experience than women. This may be consistent with the literature discussed above on the effects of exposure to technology on computer anxiety. Earlier studies found that exposure to technology was beneficial for some, but not for people with computer anxiety. Some studies have speculated that women have more computer anxiety than men. If true, it would explain the results found here. Similarly, Lips and Temple (1990) were unable to assess the effects of computer experience on womens computer usage.

Male subjects also benefited more from using computers in conjunction with a hobby, especially during elementary school and high school (Direct Factors, Appendix). Middle school and college also appear to be important time periods in which using computers with a hobby could be influential. Men may benefit more from using computers with a hobby because of their perception of computers. As was stated above, boys see computers as recreational, a toy to be played with, while girls see it as utilitarian, a tool to get a job done (Giacquinta, Bauer and E-Levin, 1993). Related to hobbies, subjects who played video games after kindergarten were more likely to be computer users than those who did not. Video games in elementary school were particularly influential for male subjects, whereas college was a much more influential time period for women (Direct Factors, Appendix). Nelson and Cooper (1989) found a similar correlation for men, but not for women.

Discussion

The goal of this study was to analyze factors that encourage people to become computer users without falling into the trap of assuming specific sets of people do not use computers. It looked at the lives of both men and women and focused on four areas which were thought to contribute to computer usage. Within each area a number of variables were found to be significantly related to being a computer user. From this type of study, however, it is difficult to tell what the relationship between the variables and computer usage is. Do the variables influence computer usage or does computer usage influence the variables? Purely quantitative research lacks important context. For example, we know that women who play video games during college are more likely to be computer users. But we do not know if women play video games during college because they are computer users, or if they are computer users because they play video games during college. Which is the cause and which is the effect? Similarly, are shy children more likely to be computer users or are computer users more likely to be shy?

This type of study does, however, display which variables deserve future investigation. For example, we know that parents influence has been overestimated by previous studies. In fact, family influences, in general, were not the most significant indicators of computer usage. Instead friends and personality and introduction and exposure to computers were far more important. Similarly, we were able to target college as an influential time in the lives of women and elementary school and middle school as influential time periods for men.

But there is much left to be investigated. Why are women most likely to be influenced during college? What are effective ways to get women involved with technology before college? If we cannot rely upon parent encouragement, how do we use the influence of peers to help encourage computer usage? Female role models appear to negatively influence middle school boys. Why? Similarly, we know very little about how the elements within each of the four areas interact with each other. Would ethnicity be significant if it were analyzed within the context of income? How do siblings affect the amount of encouragement received by parents? Whats more influential - having male role models or having male friends during elementary school?

Additionally, this study only examined four areas. There are many other factors that may influence computer usage. Schools, undoubtedly, play a significant role. Teachers, classes, friends, after-school activities are all likely to influence exposure to and attitudes about computers. Childrens play patterns, geographical location, television and sports participation, are also likely to influence computer usage. Many more factors could be enumerated.

Despite these questions, it is clear that we can learn a lot more about computer usage by focusing on computer users. Instead of discovering that people have computer anxiety and then being left with the task of wondering how to alleviate it, it makes much more sense to look at the success cases. Find out what works and then strive to replicate it. Researchers may be on their way to explaining why some people do not become computer users, but they have done little to investigate why others do. This study hopes to be a step in that direction.

Appendix

Direct Factors That May Encourage Computer Usage

Indirect Factors That May Encourage Computer Usage

References

  • Adelson, Rachel K. "Computer Confidence For Women - Daring Do For Digital Daughters." Association for Women in Computing. Online: http://www.awc-hq.org/livewire/199609.html
  • Angell, Marion D. (1991) "A Program to Develop through LOGO the computer self-confidence of seventh grade low-achieving Girls." Dissertation. ED 341382. Center for the Advancement of Education of Nova University. June.
  • Ayerman, David J and Reed, W. Michael. (1996) "Effects of Learning Styles, Programming, and Gender on Computer Anxiety" Journal of Research on Computing Education. Vol. 28 No. 2 Winter.
  • Brosnan, Mark J. (1998) Technophobia: The Psychological Impact of Information Technology. Routledge. London and New York.
  • Cassell, Justine and Jenkins, Henry. (1998) From Barbie to Mortal Kombat. The MIT Press. Cambridgem Massachusetts and London, England.
  • Clements, Douglas H. (1985) Computers in Early and Primary Education. Prentice Hall Inc. New Jersey.
  • Cockburn, Cynthia. (1983) Brothers: Male Dominance and Technological Change. Pluto Press. London.
  • Cockburn, Cynthia and Ormrod, Susan. (1993) Gender and Technology In The Making. Sage Publications. London - Thousand Oaks - New Delhi .
  • Connell, R. W.; Ashenden, D. J.; Kessler, S. and Dowsett, G. W. (1982) Making the Difference: Schools, Families and Social Division. George Allen & Unwin. Sydney, London, Boston.
  • Crook, Charles. (1994) Computers and the Collaborative Experience of Learning. Routledge. London and New York.
  • Davidson, Florence H. and Davidson, Miriam M. (1994) Changing Childhood Prejudice: The Caring Work Of the Schools. Bergin & Garvey. Westport, Connecticut.
  • "Expect The Best From A Girl" Mount Holyoke College. Online: http://www.academic.org/
  • Fuchs, Lucy. (1986) "Closing the Gender Gap: Girls and Computers" January. ED 271103
  • Gallas, Karen. (1998) Sometimes I can Be Anything: Power Gender and Identity in a Primary Classroom. Teachers College Press. Columbia University, New York and London.
  • Gattiker, Urs E. "Individual Differences and Acquiring computer Literacy: Are Women more efficient than men?"
  • Giacquinta, Joseph P.; Bauer, Jo Anne and E-Levin, Jane. (1993) Beyond Technology"s Promise. Cambridge University Press.
  • Gipson, Joella. (1997) "Girls and Computer Technology: Barrier or Key?" Educational Technology v.37 n.2 pg. 41-43, Mar. - Apr.
  • Grandy, Jerill. (1994) "Gender and Ethnic Differences Among Science and Engineering Majors: Experiences, Achievements and Expectations." GRE Board Research Report No. 92-03R. Educational Testing Service, Princeton, New Jersey. June.
  • Hawkins, Jan. (1985). "Computers and Girls: Rethinking the Issues." Sex Roles. 13: 3/4, 165 180.
  • Hughes, Martin, Brackenridge, Ann and MacLeod, Hamish. (1987) "Children's Ideas About Computers." (1987) Computers, Cognition and Development: Issues for Psychology and Education. Ed. Rutkowska, Julie C. and Cook, Charles. John Wiley & Sons. New York.
  • Jacobsdottir, Solveig and Krey, Cynthia L. (1993) "Different Computer Graphics for Girls and Boys? Preliminary Design Guidelines" January. ED 362168
  • Lips, Hilary M. and Temple, Linda. (1990) "Majoring in Computer Science: Causal Models for Women and Men." Research in Higher Education, v31, n1, pp.99-113. February.
  • Littleton, Karen and Others. (1993) "Gender and Software Effects in Computer-Based Problem Solving." March. ED 359924
  • Lloyd, Barbara and Duveen, Gerard. (1992) Gender Identities and Education: The impact of starting school. St. Martin's Press. New York.
  • Maccoby, Eleanor Emmons and Jacklin, Carol Nagy. (1974) Psychology of Sex Differences. Stanford University Press. Stanford, CA.
  • Miura, Irene and Hess, Robert D. (1983) "Sex Differences in computer Access, Interest and usage." August. ED 238400
  • Morse, Frances K. and Daiute, Colette. (1992) "I LIKE computers versus I LIKERT computers: Rethinking Methods for Assessing the Gender Gap In Computing." Paper presented at the Annual Conference of the American Educational Research Association (San Francisco, CA, April 20-24, 1992). ERIC ED349939
  • Murphy, Vaugn and Thuente, Kimberly. (1995) "Using Technology in Early Learning Classrooms" Learning and Leading with Technology: The ISTE Journal of Educational Technology Practice and Policy. Vol. 22 No. 8 May, pg. 5
  • "New Study Shows Preteens Are Biggest Computer Users" The Roper Youth Report. The Global Schoolhouse Online: http://www.gsn.org/wce/archives/heller6.htm
  • Nelson, Lori J. and Cooper, Joel. (1989) "Sex Role Identity, Attributional Style, and Attitudes toward computers." 31 March. ED 314658
  • Nix, Don and Spiro, Randy. (1990) Cognition, Education, Multimedia: Exploring Ideas in High Technology. Lawrence Erlbaum Associates, New Jersey.
  • Sadker, Myra and David. (1994) Failing at Fairness: How America's Schools Cheat Girls. Charles Scribner's Sons. New York.
  • Sanders, Jo. (1993) "Closing The Gender Gap." Executive Examiner. September. pg. 32
  • Sanders, Jo and Stone, Antonia. (1986) The Neuter Computer. Neal - Schuman Publishers Inc., New York.
  • Schmitt, Dorren Rafael. (1990) Curriculum Implications for Using Computers with Pre-Readers - What are the socializing aspects? University of New Orleans and Edcom. November. ED 326315
  • Shaw, Jenny. (1995) Education, Gender and Anxiety. Taylor & Francis, London.
  • Singh, Jagjit K.(1992) "Attitudes towards, Preferences for and Knowledge of Computers Among Gifted and Average Junior High Children." September. Ed 371550
  • Spender, Dale. (1995) Nattering On The Net. Garamount Press.
  • Stalker, Sylvia. (1983) "Computers in the Classroom: A Feminist Issue" June. ED 240015
  • Sturm, Deborah and Moroh, Marsha. (1995) "Gender and Computer Science Majors: Perceptions and Reality."
  • Swanson, Janese. "What's the Difference?" Girltech Online: http://www.girltech.com/
  • Swoope, Karen and Johnson, Carole Schultz. (1985) "Students Perceptions of Interest in Using Computers: Boys, Girls, or Both?" April. Ed 263899
  • Thorne, Barrie. (1993) Gender Play. Rutgers University Press, New Brunswick, New Jersey.
  • Volman, Monique. (1995) Gender Related Effects of Computer and Information Literacy Education ED 382188 Netherlands
  • Wajcman, Judy. (1991) Feminism Confronts Technology. Pennsylvania State University Press.
  • Wilkinson, Alex Cherry. (1983) Classroom Computers and Cognitive Science. Academic Press, New York.
  • Witkin, Herman A. and Goodenough, Donald R.(1981) Cognitive Styles: Essence and Origins: Field Dependence and Field Independence. International Universities Press, Inc. New York.