Internet Self-Efficacy,the Need for Cognition,and Sensation Seeking as Predictors of Problematic Use of the Internet

Abstract

We conducted a study of the relationships between Internet self-efficacy, sensation seeking, the need for cognition, and problematic use of the Internet. The study was based on a randomly selected sample of 979 adult Internet users. Hierarchical multiple regression analysis of these subjects' responses on a questionnaire consisting of relevant items indicated that Internet self-efficacy and sensation seeking positively predicted problematic Internet use. Contrastingly, the need for cognition was significantly negatively associated with problematic Internet use.

Introduction

The Internet has developed rapidly and is increasingly popular on a worldwide scale. The China Internet Network Information Center (CNNIC) reported in July 2009 that in China alone, 338 million people had access to the Internet.¹ Although the Internet has proven highly beneficial, its potential for adverse or problematic use has also been recognized, and considerable research has recently focused on studying its individually problematic, unhealthy, and pathological use.^2–4 Studies of this have shown that only certain types of individuals, such as those who are lonely, depressed, or homesick, are vulnerable to adverse use of the Internet.^3–5 We conducted a study to examine the predictiveness of self-efficacy in use of the Internet, sensation seeking, and the need for cognition to problematic use of the Internet.

Bandura defined self-efficacy as the belief in one's ability to perform a given task.⁶ According to Social Cognition Theory, self-efficacy may have an influence on one's effort, persistence, and expressed interest in performing a task or achieving a goal of a particular level of difficulty.⁷ As a new technology, the Internet requires users to acquire certain skills, and some Internet applications can be viewed as tasks.⁸ Internet self-efficacy may therefore exert an effect on individuals' Internet usage. Eastin and LaRose found greater Internet self-efficacy among people who used the Internet more frequently,⁷ and Whitty and McLaughlin found that people with greater Internet self-efficacy were more likely to use the Internet for entertainment.⁹ Because specific skills and knowledge are needed to perform Internet-related tasks (e.g., playing online games, downloading videos, and using software provided via the Internet), people with low Internet self-efficacy often feel that these tasks are too difficult for them and avoid using such Internet applications. On the other hand, people with high Internet self-efficacy are free of such problems but are more vulnerable to potentially addictive applications of the Internet such as online games.

Sensation seeking is defined as a personal trait characterized by a “need for varied, novel, and complex sensations and experiences and the willingness to take physical and social risks for the sake of such experiences.”¹⁰ Individuals with greater levels of sensation seeking express greater than average intolerance for boredom and seek more than average external stimulation, and are therefore more susceptible to addictive behaviors.¹¹ Lin and Tsai suggest that many online activities, including surfing the Internet, are widely regarded as venues for sensation seeking. They found that Internet-dependent individuals scored significantly higher on overall measures of sensation seeking than individuals not dependent on the Internet.¹² Hence, we propose that people with higher sensation seeking are more vulnerable to problematic Internet use.

The need for cognition is referred to as one's tendency to engage in and enjoy effortful handling of information and cognitive endeavors.¹³ Individuals with a high need for cognition are viewed as having a natural and intrinsic motivation to seek knowledge.¹⁴ Kaynar and Amichai-Hamburger recently found that people with greater levels of need for cognition are more likely to use the Internet for professional purposes, such as sending work-related e-mail, using real-time messaging in performing work, and obtaining study-related information.¹⁵ The third hypothesis for our study was therefore that individuals with a high need for cognition may use the Internet in a constructive and healthy way rather than detrimentally, such as in an addictive dependence on leisure entertainment.

The current study aims to provide the preliminary evidence for the effects of Internet self-efficacy, sensation seeking, and need for cognition in predicting problematic Internet use. After the forgoing reasoning, we propose:

H1: Adult users' Internet self-efficacy level is a significant positive predictor of their level of problematic Internet use.

H2: Sensation seeking could positively predict one's problematic Internet use level.

H3: Need for cognition is a significant negative predictor of problematic Internet use.

Methods

Participants

We invited a random sample of adult passersby to participate in our study in Beijing and Shenzhen, two major cities in China. A total of 1,200 volunteers agreed to participate in our study and filled out the survey. However, 221 of the 1,200 surveys were not completed and were thus not usable in subsequent analysis, resulting a final sample of 979 adults (response rate = 81.58%). About half of the participants were male (49.03%). The average age of this sample was 25.84 (SD = 5.86). On average, participants had used the Internet for 5.43 years (SD = 3.16).

Measures

Using a translation and back-translation procedure described by Brislin,¹⁶ the original English versions of the Internet Self-Efficacy,¹⁷ Sensation Seeking,¹⁸ and Problematic Internet Use¹⁹ scales were first translated into Chinese by Chinese undergraduate students majoring in psychology and, independently, by a similar group of graduate students. Discussion among the two groups of translators yielded a Chinese version of these scales. A certified Chinese–English translator then translated these Chinese versions back into English, comparing the English translations of each scale with the original English-language scales and revising the Chinese versions accordingly. To produce the best translation for items for which the back-translated English version disagreed with the original scale, we consulted two psychology professors who were bilingual in Chinese and English. Through this, we established the final Chinese translations of each scale.

Internet self-efficacy

Internet self-efficacy was measured by the Internet Self-Efficacy Scale developed by Torkzadeh and Van Dyke.¹⁷ The scale consists of 15 items. Participants were instructed to score the extent of the confidence they felt about each item on a 7-point scale, where 1 = “strongly disagree” and 7 = “strongly agree”. A high score means a high level of Internet self-efficacy. The translation scale displayed adequate internal consistency in this study (Cronbach's α = 0.92).

Need for cognition

In contrast to Internet self-efficacy, sensation seeking, and problematic use, which were assessed with the translated and back-translated scales described above, need for cognition was assessed with the Chinese version of the Need for Cognition Scale developed by Kuang et al.,²⁰ which contains 18 items. Participants responded to each item by describing themselves on a 7-point scale, where 1 = “strongly disagree” and 7 = “strongly agree”. A high score indicates a high level of need for cognition. The Chinese version has previously displayed good reliability and validity,²⁰ and it displayed adequate internal consistency (α = 0.80) in the current sample.

Sensation seeking

Sensation seeking was measured with the Brief Sensation Seeking Scale developed by Hoyle et al.,¹⁸ which includes eight items. Participants were instructed to score the extent to which they agreed with each item according to a 5-point scale, where 1 = “strongly disagree” and 5 = “strongly agree”. A high score indicates that the participant has a high level of sensation seeking. In this study, the Cronbach's alpha of the translation scale was 0.65.

Problematic Internet use

Problematic Internet use was assessed with the Problematic Internet Use Scale developed by Jia and Jia,¹⁹ which consists of 10 items. Participants responded to each item by describing themselves on a 7-point scale, where 1 = “strongly disagree” and 7 = “strongly agree”. High scores on this scale suggest high levels of problematic Internet use. In this study, the translation scale had a good reliability coefficient (α = 0.87).

Data analysis

Analysis of the data was done through hierarchical multiple regression analysis, with problematic Internet use as the dependent variable. In the first step of this process, gender and age were entered as demographic predicate variables. Following this, Internet self-efficacy, need for cognition, and sensation seeking were entered after controlling for the effects of the demographic variables.

Results

Means, standard deviations, and bivariate correlations of all study variables are reported in Table 1. The correlations displayed in Table 1 show that the levels of Internet self-efficacy and sensation seeking were significantly and positively correlated with problematic Internet use, and need for cognition was negatively related to problematic Internet use. In addition, gender was positively associated with problematic Internet use.

Table 1.

Means, Standard Deviations, and Correlations Among Study Variables

Variable	M	SD	1	2	3	4	5	6
1. Gender	—	—	—
2. Age	25.84	5.86	0.09^**	—
3. Internet self-efficacy	5.16	1.08	0.11^***	−0.05	—
4. Need for cognition	4.46	0.79	0.18^***	−0.11^***	0.30^***	—
5. Sensation seeking	2.96	0.72	0.09^**	−0.14^***	0.18^***	0.12^***	—
6. Problematic Internet use	3.71	1.15	0.07^*	−0.02	0.13^***	−0.19^***	0.23^***	—

Note. N = 979. Gender is coded as “1” for male and “0” for female.

p < 0.05; **p < 0.01; ***p < 0.001.

To examine the effects of Internet self-efficacy, need for cognition, and sensation seeking whilst considering the effects of other variables in predicting problematic Internet use, hierarchical multiple regression analysis was conducted. The results in Table 2 show that after controlling for age and gender, Internet self-efficacy, need for cognition, and sensation seeking all significantly predicted the levels of problematic Internet use. This integrative model is significant, F(5, 973) = 30.51, p < 0.001. Gender was significantly related to problematic Internet use, β = 0.08, p < 0.05. Internet self-efficacy, need for cognition, and sensation seeking together explained an additional 13% of the variance of problematic Internet use after controlling for gender and age, F_change(3, 973) = 49.04, p < 0.001. Internet self-efficacy and sensation seeking were significant predictors of problematic Internet use, β = 0.17, p < 0.001 and β = 0.23, p < 0.001 respectively, such that individuals with higher levels of Internet self-efficacy or individuals with higher levels of sensation seeking were more likely to be problematic Internet users. Need for cognition level significantly and negatively predicted one's problematic Internet use, β = − 0.28, p < 0.001, indicating that individuals with lower levels of need for cognition were more likely to be problematic Internet users. And results suggested that there was no collinearity between predictors in the regression equations. All three hypotheses were supported.

Table 2.

Hierarchical Multiple Regression Analyses in Predicting Levels of Problematic Internet Use on the Basis of Internet Self-Efficacy, Need for Cognition, and Sensation Seeking

Step	Variable	B	SE	β	t	R ²	ΔR ²	VIF
1	Constant	3.75	0.17		22.37^***
	Gender	0.16	0.07	0.07	2.11^*			1.01
	Age	0.00	0.01	−0.02	−0.72	0.01	0.01	1.01
2	Constant	3.52	0.32		11.17^***
	Gender	0.18	0.07	0.08	2.51^*			1.06
	Age	0.00	0.01	−0.01	−0.49			1.05
	Internet self-efficacy	0.18	0.03	0.17	5.30^***			1.12
	Need for cognition	−0.41	0.05	−0.28	−8.93^***			1.14
	Sensation seeking	0.37	0.05	0.23	7.48^***	0.14^***	0.13^***	1.06

Note. N = 979. B, unstandardized regression coefficients; SE, standard error of B's mean; β, standardized regression coefficients; t, value of t test for β; R² = variance of independent variable that could be explained by the predictors/the total variance of independent variable; VIF, Variance Inflation Factor, the indictor of collinearity.

p < 0.05; ***p < 0.001.

Discussion

The study described here primarily examined the predictiveness of Internet self-efficacy, sensation seeking, and need for cognition to problematic Internet use. Our findings generally support the hypothesized relationships of the first three of these variables to the last variable, as given earlier. The study expands the understanding of the personal characteristics related to detrimental use of the Internet.

Specifically, levels of Internet self-efficacy positively predicted problematic Internet use. This result is consistent with the finding in a previous study that Internet self-efficacy is related to use of the Internet for entertainment.⁹ This finding can be interpreted as covering many Internet applications, with those that are addictive especially requiring users to master specific skills, whereas users who are not confident of their Internet-related skills will find such Internet applications especially daunting. The conclusion that can be drawn from this is that users with higher levels of Internet self-efficacy and the ability to handle complex Internet requirements easily are at the same time susceptible to problematic use of the Internet.

Our findings may shed light on the difference between general self-efficacy and task-specific self-efficacy. General self-efficacy is often shown to be related to positive outcomes such as high performance, psychological well-being, and job satisfaction. However, as revealed in our study, Internet self-efficacy, as a type of task-specific self-efficacy, is associated with negative impact of Internet. This may be attributed to the fact that general self-efficacy concerns one's perception about one's ability to perform tasks, broadly including one's task-related skills and broadly defined ability such as self-control, whereas Internet self-efficacy focuses on one's confidence of specific capacity related to Internet use, such as browsing skills. The Internet-related tasks were also inherently different from general tasks in that the former might be risky inconspicuously.

Our study also showed that sensation seeking could predict problematic Internet use. This is in consistent with the finding that sensation seeking is positively correlated with risky and addictive behaviors such as smoking, use of alcohol, and use of illicit drugs.^21–23 This result is reasonable because the Internet provides a constantly novel environment with a plenitude of opportunities for adventure. Accordingly, users with greater levels of sensation seeking are likely to be fascinated by the Internet and suffer the adverse effects of its use.

As expected, the need for cognition was negative and a significant predictor of problematic Internet use. This is congruent with the previous finding that the need for cognition is positively associated with use of the Internet for work or study.¹⁶ Our result may be interpreted as showing that individuals with higher levels of need for cognition use the Internet in a healthy and controllable way for professional purposes, and are therefore more removed from problematic use of the Internet.

Besides, gender as a control variable is significantly related to problematic Internet use. This finding is consistent with a previous study that male users may be more vulnerable to the detrimental impact of the Internet.³ This could be explained by the fact that some males have problems with developing intimate relationships, so they prefer to communicate via the Internet³ and play online games to kill their loneliness.

As a preliminary exploration, our study has some limitations. The first one concerns the sample of the current study; the randomly selected sample might be biased due to the specific location of sampling. Further study is needed to replicate our research in other samples. Another limitation of the study concerns the cross-sectional design. Regression cannot confirm the direction of causality, and the relationships could be explained in both directions. Longitudinal research is needed to confirm further the causal relationships hypothesized in this research. Finally, we used the translation version of scales in this study, and the Cronbach's alpha of the translation scale of sensation seeking was not particularly high (0.65). There might be cultural differences in the way many Chinese would respond to some of these items. Maybe a Chinese person would prefer to hide a negative criticism of him/her to be a sensation seeker. Although most of the translation scales manifest good reliabilities and the original versions have adequate validities, further research is needed to substantiate our findings by examining the validities and reliabilities of these translation scales.

Further study is needed of the variables that may link Internet self-efficacy, sensation seeking, the need for cognition, and problematic use of the Internet. This may provide a better understanding of the mechanisms that connect these personal variables with adverse effects of Internet use.

In conclusion, our research, although preliminary, emphasizes the importance of several personal factors such as Internet self-efficacy, sensation seeking, and the need for cognition in problematic use of the Internet. Particular attention should be given to these characteristics, and users in whom they represent risk factors should be protected against detrimental effects of the Internet. We hope that our research will promote further investigation of the psychological and behavioral characteristics that predict problematic use of the Internet, and contribute to effective strategies for preventing it.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

China Internet Network Information Center (CNNIC). 2009. Statistical report of Internet development of China. http://research.cnnic.cn/img/h000/h11/attach200907161303030.pdf. 2009 Dec. 7.

Ozcan

, Buzlu

. Internet use and its relation with the psychosocial situation for a sample of university students. CyberPsychology & Behavior, 2007; 10:767–72.

Ceyhan

. Predictors of problematic Internet use on Turkish university students. CyberPsychology & Behavior, 2008; 11:363–6.

, Yan

, Chen

et al. Factors influencing Internet addiction in a sample of freshmen university students in China. CyberPsychology & Behavior, 2009; 12:327–32.

Ceyhan

, Ceyhan

. Loneliness, depression, and computer self-efficacy as predictors of problematic Internet use. CyberPsychology & Behavior, 2008; 11:699–701.

Bandura

. Self-efficacy: toward a unifying theory of behavioural change. Psychologist, 1977; 37:122–47.

GIST

. Self-efficacy: Implications for organizational behavior and human resource management. Academy of Management Review, 1987; 12:472–86.

Eastin

, LaRose

. Internet self-efficacy and the psychology of the digital divide. Journal of Computer-Mediated Communication, 2000; 6. www.ascusc.org/jcmc/vol6/issue1/eastin.html. 2009 Dec. 7.

Whitty

, McLaughlin

. Online recreation: The relationship between loneliness, Internet self-efficacy and the use of the Internet for entertainment purposes. Computers in Human Behavior, 2007; 23:1435–46.

10.

Zuckerman

. 1979. Sensation seeking: Beyond the optimal level of arousal. Hillsdale, NJ: Lawrence Erlbaum.

11.

Pizam

, Jeong

G-H

, Reichel

et al. The relationship between risk-taking, sensation seeking, and tourist behavior of young adults: A cross-cultural study. Journal of Travel Research, 2004; 42:251–60.

12.

Lin

SSJ

, Tsai

. Sensation seeking and Internet dependence of Taiwanese high school adolescents. Computers in Human Studies, 2002; 18:411–26.

13.

Cacioppo

, Petty

. The need for cognition. Journal of Personality & Social Psychology, 1982; 42:116–31.

14.

Verplanken

, Hazenberg

, Palenéwen

. Need for cognition and external information search. Journal of Research in Personality, 1992; 26:128–36.

15.

Kaynar

, Amichai-Hamburger

. The effects of cognition on Internet use revisited. Computers in Human Behavior, 2008; 24:361–71.

16.

Brislin

. Back-translation for cross-cultural research. Journal of Cross-Cultural Psychology, 1970; 1:185–216.

17.

Torkzadeh

, Van Dyke

. Development and validation of an Internet self-efficacy scale. Behaviour & Information Technology, 2001; 20:275–80.

18.

Hoyle

, Stephenson

, Palmgreen

et al. Reliability and validity of a brief measure of sensation seeking. Personality & Individual Differences, 2002; 32:401–14.

19.

Jia

, Jia

. Factorial validity of problematic Internet use scales. Computers in Human Behavior, 2009; 25:1335–42.

20.

Kuang

, Shi

, Cai

. The Chinese version of Need for Cognition Scale. Chinese Mental Health Journal, 2005; 19:57–60.

21.

Zuckerman

, Ball

, Black

. Influences of sensation, gender, risk appraisal, and situational motivation on smoking. Addictive Behaviors, 1990; 15:209–20.

22.

Stacy

, Newcomb

, Bentler

. Cognitive motivations and sensation seeking as long-term predictors of drinking problems. Journal of Social & Clinical Psychology, 1993; 12:1–24.

23.

Newcomb

, McGee

. Influence of sensation seeking on general deviance and specific problem behaviors from adolescence to young adulthood. Journal of Personality & Social Psychology, 1991; 61:614–28.