Skip to content

Advertisement

  • Research note
  • Open Access

Psychometric properties of a Thai version internet addiction test

BMC Research Notes201811:69

https://doi.org/10.1186/s13104-018-3187-y

  • Received: 21 November 2017
  • Accepted: 17 January 2018
  • Published:

Abstract

Objective

The aim was to assess the reliability and validity of a Thai version internet addiction test.

Results

Cronbach’s alpha for the Thai version of the internet addiction test was 0.89. A three-factor model showed the best fit with the data for the whole sample, whereas the hypothesized six-factor model, as well as a unidimensional model of the internet addiction test, failed to demonstrate acceptable fit with the data. Three factors, namely functional impairment, withdrawal symptoms and loss of control, exhibited Cronbach’s alphas of 0.81, 0.81, and 0.70, respectively. Item 4, ‘to form new relationships with online users’, yielded the lowest loading coefficient of all items. Positive correlations between the internet addiction test and UCLA loneliness scores were found. The Thai version of the internet addiction test was considered reliable and valid, and has sufficient unidimensionality to calculate for total score in screening for excessive internet use.

Keywords

  • Internet
  • IAT
  • Loneliness
  • Psychometrics
  • Validity
  • Thai

Introduction

Internet addiction has been specified as pathological internet use, compulsive Internet use and problematic Internet use [1, 2]. The inability to control Internet use impacts the body’s immune system, and work, social and academic performance [35]. Factors associated with Internet addiction include social isolation, low self-esteem, depression, personality traits, and interpersonal problems [68]. Despite the fact that problems arise from pathological use of the Internet, Internet addiction has not yet been included in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). The only behavioral addiction included in the latest version of the DSM is gambling disorder, which was included in a section recommending further study.

The prevalence of Internet addiction is estimated to be between 5.2 and 80.5% [9]. In Thailand, internet addiction has been investigated in both college and clinical settings, and 24.4% of participants were found to exhibit a low level of internet addiction [10, 11].

A number of instruments have been used to assess Internet addiction. The internet addiction test (IAT), developed by Young, is one of the most common tools and has been widely tested for its psychometric properties [1216]. The IAT has demonstrated clinical relevance to addictive properties such as problems with time management, loss of control, impact on relationships [17]. The IAT has been shown to have excellent reliability, as assessed by internal consistency using Cronbach’s alpha, and good test–retest reliability [18]. Two studies from Korea, and one from Germany reported 2-week correlations ranging from 0.77 to 0.85 [1820].

The validity of the test has been examined and reported in several languages including French [21], Italian [22], Portuguese [23], Spanish [16], German [24], Chinese [14], Malay [25], and Vietnamese [26]. The number of factors discovered through related research is inconsistent, ranging from 1 to 6 factors, depending on the sample and study setting employed [21, 22, 25, 27, 28]. Young proposed the six-factor model in the early stages of investigation of the IAT [28]. However, in a subsequent study with a larger sample size, the three-factor model was revealed to fit the data better [29]. Recent studies have found that two factors can adequately represent the model [12, 22]. The item commonly found to be misfitted is item 4, ‘to form new relationships with online users’, while other misfit items vary from study to study [18, 22].

While the IAT is increasingly being used to examine the Thai population, the Thai version of the IAT has never been examined for validity and reliability. The purpose of the present paper is to investigate the psychometric properties of the Thai version of the IAT in terms of factor structure, internal consistency, and possible misfit item(s) with other studies, as well as examining concurrent validity with other measurements.

Main text

Methods

Participants

First to sixth–year medical students from Chiang Mai University, Thailand, participated in this study by means of convenience sampling in 2015. Of 324 participants, 56.8% were female. The mean age was 20.88 (SD 1.8). All completed demographic data questionnaires and answered questions regarding their internet use using the IAT and the UCLA loneliness scale. The average time spent on the internet daily was 4.9 h (SD = 2.7), and 6.9 days per week. According to Young’s addiction level, 63.3%, none; 30.9%, mild; 5.2%, moderate; and 0.6%, severe.

Measurements

The Thai version of the IAT: The IAT, developed by Young [17], is a 20-item self-report instrument in which respondents rate their tendency for addiction to the internet using a five-point scale, ranging from 1 (rarely) to 5 (always). The IAT total score is the sum of the ratings given by the examinee in response to the 20 items. The higher the score, the higher the level of severity of internet compulsivity and addiction. Total scores of more than 30 are considered to indicate addiction. We obtained permission from Dr. Kimberly Young to translate the IAT and use it in our research study. The forward and backward translation process was done. The study sample had a Cronbach’s alpha of 0.89.

UCLA loneliness scale: The UCLA loneliness scale is an assessment tool used to screen and assess the severity of loneliness. It consists of 20 items asking respondents how often they experienced feelings of loneliness during the past week. Four-point responses to each item range from 0 (‘not at all’) to 3 (‘almost daily’). The Thai version demonstrated good reliability and validity [30]. The UCLA loneliness scale was used to evaluate the convergent validity of the Thai version of the IAT because a correlation between loneliness and internet addiction has been previously reported [31, 32].

Statistical analysis

Descriptive statistics was used for demographic data as well as data screening analysis for factor analysis. Item responses exhibited skewness and kurtosis (> ± 2) [33]. Exploratory factor analysis (EFA) using ordered categorical (ordinal) response was performed. For parameter estimation, as data were ordinals, robust weighted least square means and variance adjusted (WLSMV) were employed for estimators [34]. Confirmatory factor analysis (CFA) was used to test Young’s hypothesized six-factor model [28].

Regarding fit indexes, in the Comparative Fit Index (CFI) and Tucker–Lewis Index (TLI), a value > 0.90 indicates reasonable fit, and a value > 0.95 indicates good model fit; a weighted root-mean-square residual (WRMR) lower than 0.9, a root-mean-square error of approximation (RMSEA) of ≤ 0.06, and a standardized root mean square residual (SRMR) of < 0.08 indicate a reasonable fit [3537]. In addition, the χ2 statistic has been used to test the goodness of model fit if the ratio χ2/df < 3 [33]. Missing data (0.5%) were replaced by series of means. Modification indices were added to the model after the initial analysis, and CFA was carried out using Mplus 7.4 [38]. Pearson’s correlation analyses were used to determine concurrent validity, and Cronbach’s alpha was calculated to assess internal consistency.

Results

The objective of the analyses was to examine the internal consistency and the factor structure of the translated Thai version of the IAT.

Table 1 shows the mean and SD of each item. Cronbach’s alpha for all items was 0.89. The Corrected Item-Total Correlations were > 0.4, except for items 4 and 7, for which the alpha was estimated to be higher when either of these two items was deleted.
Table 1

Mean, SD, Cronbach’s alpha, and item-total correlation among IAT items

 

Mean

SD

Corrected item-total correlation

Squared multiple correlation

Cronbach’s alpha if item deleted

1. Stay online longer than intended

3.25

0.996

0.438

0.343

0.884

2. Neglect household chores

2.41

1.149

0.496

0.407

0.883

3. Prefer excitement of internet

1.81

1.051

0.509

0.328

0.882

4. New relationships with online users

1.59

1.188

0.313

0.179

0.888

5. Others complain to you

1.2

1.052

0.553

0.361

0.881

6. Your work suffers

1.25

1.134

0.542

0.425

0.881

7. Check email before something else

1.03

1.23

0.363

0.235

0.887

8. Your job performance suffers

1.4

1.054

0.578

0.461

0.88

9. Become defensive when asked

0.74

0.918

0.496

0.351

0.882

10. Block out disturbing thoughts

1.09

1.128

0.536

0.345

0.881

11. Find yourself anticipating

1.56

1.104

0.563

0.417

0.881

12. Fear life without the internet

1.72

1.344

0.439

0.357

0.885

13. Snap if someone bothers

0.46

0.808

0.52

0.372

0.882

14. Lose sleep due to being online

2.24

1.289

0.522

0.353

0.882

15. Feel preoccupied with the internet

0.67

0.964

0.56

0.444

0.881

16. “Just a few more minutes”

1.73

1.444

0.482

0.382

0.884

17. Try to cut down the amount of time

1.78

1.231

0.583

0.479

0.88

18. Try to hide how long

0.67

0.946

0.56

0.377

0.881

19. Prefer spending time online

1.49

1.235

0.505

0.317

0.882

20. Feel depressed, when off-line

0.75

1.03

0.528

0.42

0.881

The eigenvalues for the sample correlation matrix (EFA) were 7.811, 1.723, and 1.269. The factor loading coefficients are set out in detail in Table 2. The scree plot suggested three factors due to the manner in which the slope levels off. The scree plot according to eigenvalues (Additional file 1: Figure S1) determined the number of these three factors (components).
Table 2

Cronbach’s alpha and loading coefficients of one-factor and 3-factor solutions of the Thai version of the IAT

IAT item

One-factor

Component

F1_functional impairment

F2_withdrawal symptom

F3_loss control

8. Your job performance suffers

0.678

0.743

0.449

0.283

6. Your work suffers

0.643

0.717

0.391

0.330

5. Others complain to you

0.628

0.658

0.497

0.099

2. Neglect household chores

0.629

0.650

0.359

0.419

9. Become defensive when asked

0.598

0.630

0.531

− 0.087

10. Block out disturbing thoughts

0.628

0.619

0.546

0.056

1. Stay online longer than intended

0.538

0.549

0.282

0.468

3. Prefer excitement of internet

0.605

0.546

0.496

0.240

7. Check email before something else

0.439

0.472

0.335

0.079

15. Feel preoccupied with the internet

0.694

0.446

0.778

0.080

20. Feel depressed, when off-line

0.652

0.461

0.750

− 0.070

13. Snap if someone bothers you

0.655

0.474

0.719

0.003

11. Find yourself anticipating

0.652

0.436

0.711

0.114

18. Try to hide length of internet use

0.668

0.538

0.666

0.075

12. Fear life without the internet

0.520

0.297

0.647

− 0.043

19. Prefer spending time online

0.583

0.499

0.562

0.052

14. Lose sleep due to being online

0.608

0.436

0.538

0.365

4. New relationships with online users

0.359

0.364

0.390

− 0.198

17. Try to cut down the amount of time

0.700

0.513

0.535

0.599

16. “Just a few more minutes”

0.589

0.364

0.472

0.592

Cronbach’s alpha

0.89

0.81

0.81

0.70

Cross-loadings (≥ 0.4) were found in all items except for items 6, 8 and 12. Item 4 had the lowest loading coefficient (0.356), followed by item 7 (0.439) on the one-factor model. Factor 1 covered items describing performance problems at school or at work, and relationship problems due to excessive Internet use. Factor 2, which dealt with ‘withdrawal symptoms’, involved depression, irritability, and fear when not using the Internet. Factor 3, ‘loss of control’, contained items about failed attempts to cut down on Internet use. The correlation coefficients between Factors 1 and 2, 1 and 3 and 2 and 3 were 0.594 (p < .05); 0.175 (p > .05) and 0.014 (p > .05), respectively. The Cronbach’s alphas for the factors were 0.70–0.81, and 0.89 for all items. The total scores of the Thai version of the IAT were significantly correlated with the UCLA loneliness scale (r = 0.293, p < .001). Level of loneliness, which is related to Internet addiction, was also significantly correlated, which provides strong support for the concurrent validity of the Thai version of the IAT.

Table 3 shows the goodness of fit of various models. The unidimensional models demonstrated poor fit, whereas the hypothesized six-factor models yielded better fit but were still unacceptable. The three-factor model provided the best fit to the data, with CFI and TLI values exceeding 0.95 and an RMSEA of less than 0.06.
Table 3

Comparison of fit indices among three proposed models

Fit indices

χ2

Df

p value

Chi/df

RMSEA

CFI

TLI

SRMR

Hypothesized-6 factor

521.019

155

0.000

3.361

0.085 (0.077–0.094)

0.903

0.881

1.254

Three-factor

248.516

133

0.000

1.869

0.052 (0.042–0.062)

0.969

0.956

0.044

One-factor

565.112

170

0.000

3.324

0.085 (0.077–0.092)

0.895

0.883

0.079

df degrees of freedom, RMSEA root-mean-square error of approximation, CFI comparative fit index, TLI Tucker-Lewis index, SRMR standardized root mean square residual

Discussion

This study found that the Thai version of the IAT demonstrates similar satisfactory results to the previous study, particularly among the medical student sample. The corrected R squared of item 4 (‘forming new relationships with fellow on-line users’) presented the lowest value (0.179) of all items, which endorsed the findings of related studies in different cultures [12, 18, 22]. We hypothesized that this item may reflect the pathological behaviors associated with excessive Internet use less accurately than the remainder of the items.

In exploring factor structure, our results extracted three factors, whereas other researchers have proposed various factor solutions, ranging from 1 to 6. The first analysis of the IAT by two studies yielded a six factor-solution [28, 39]. Notably, both studies analyzed the test using a relatively low sample size (a sample of 86 in the former and 115 in the latter study). When repeated with a larger sample size, the number of factors was generally reduced to 2–3 [20, 22, 40]. This difference was likely due to the differences among the samples, which variously comprised college students and clinical samples, and whose cultural backgrounds differed.

Our findings differed from that of other research conducted among medical students. A Malay sample identified a five-factor model of the IAT [25], whereas a three-factor solution was found among Greek, Persian, and Pakistani samples [4143]. Except for Malaysia, samples comprising medical students provided the same three-factor solution in various studies. Items within each factor differed due to the differences between the various samples’ characteristics, environments, and cultural and religious backgrounds.

Because no hypothesized model was proposed by Young in the original version of the IAT, attempts have been made to design more valid and reliable items in a newer version. In revising the IAT, many investigators have suggested removing the item(s) with low loading coefficients on designated factors or reducing problems with cross-loadings on other factors. However, these suggestions may be of limited relevance to the studied samples, and cannot necessarily be applied to other studies. One important property of the scale is its unidimensionality, which requires creating a cut-off score to define various levels of the problem. The unidimensionality of the model should be warranted to determine the legitimacy of summarizing the scale in a single score [44, 45]. Some investigators have suggested that, when the ratio of first-to-second eigenvalues is greater than four, the model can be considered unidimensional [46, 47].

To date, we have found only two studies identifying the IAT as a unidimensional model [13, 21]; one seems to show that it has sufficient unidimensionality [27]. The remainder of the existing studies found two or more factor solutions, including the present study. However, using the above criteria, the Thai version of the IAT showed sufficient unidimensionality. In addition, in the revising process, culturally biased items and differential item functioning due to sex should be identified to make the test more capable of being compared across cultures.

In conclusion, the Thai version of the IAT was shown to have good reliability and concurrent validity, as demonstrated by a significant correlation between the UCLA loneliness scale and the IAT. This relationship reflects some convergent validity because loneliness and Internet addiction were found to be correlated, especially among university or college students [31, 32]. A three-factor model fits the data well; in addition, it has sufficient unidimensionality to allow the total score to be used to screen for excessive Internet use.

Limitations

This study has some limitations. First, even though the sample represents all levels of medical student, recruitment was not randomized. Second, test–retest reliability was not conducted to ensure temporal stability. Third, medical students may not be representative of the general population, so further examination of the validity and reliability of the test should be conducted in other populations. Further exploration in relation to shortening the scale to make it more unidimensional is encouraged.

Abbreviations

CFA: 

confirmatory factor analysis

CFI: 

Comparative Fit Index

DSM-5: 

diagnostic and statistical manual of mental disorders

EFA: 

exploratory factor analysis

IAT: 

internet addiction test

RMSEA: 

root-mean-square error of approximation

SRMR: 

standardized root mean square residual

TLI: 

Tucker–Lewis Index

UCLA: 

University of California Los Angeles

WLSMV: 

weighted least square means and variance adjusted

WRMR: 

weighted root-mean-square residual

Declarations

Authors’ contributions

AN, MP, NW and TW conceived the study. AN, MP, NW and TW designed the study. SS and PK collected and cleansed data. NW and TW analyzed the data. AN, MP, NW and TW wrote the initial draft. AN, MP, SS, PK, NW and TW revised the draft manuscript. All authors read and approved the final manuscript.

Acknowledgements

The authors thank all participating medical students, who made this study successful.

Competing interests

The authors declare that they have no competing interests.

Availability of data and materials

The datasets generated and/or analyzed during the current study are available from the authors on reasonable request.

Consent for publication

Not applicable.

Ethics approval and consent to participate

Ethical approval for this study was obtained from the Ethics Review Committee, Faculty of Medicine, Chiang Mai University (Ethics Approval Number REC-25580428-04268). Informed written consent was obtained from all participants.

Funding

Not applicable.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Authors’ Affiliations

(1)
Department of Psychiatry, Faculty of Medicine, Chiang Mai University, 110 Intawaroros Rd., T. Sriphum, A. Muang, Chiang Mai, 50200, Thailand

References

  1. Greenfield DN. Psychological characteristics of compulsive internet use: a preliminary analysis. Cyberpsychol Behav. 1999;2:403–12.View ArticlePubMedGoogle Scholar
  2. Shapira NA, Goldsmith TD, Keck PE, Khosla UM, McElroy SL. Psychiatric features of individuals with problematic internet use. J Affect Disord. 2000;57:267–72.View ArticlePubMedGoogle Scholar
  3. Engelberg E, Sjöberg L. Internet use, social skills, and adjustment. Cyberpsychol Behav. 2004;7:41–7.View ArticlePubMedGoogle Scholar
  4. Leung L, Lee PSN. Impact of internet literacy, internet addiction symptoms, and internet activities on academic performance. Soc Sci Comput Rev. 2012;30:403–18.View ArticleGoogle Scholar
  5. Reed P, Vile R, Osborne LA, Romano M, Truzoli R. Problematic internet usage and immune function. PLoS ONE. 2015;10:e0134538.View ArticlePubMedPubMed CentralGoogle Scholar
  6. Zamani BE, Abedini Y, Kheradmand A. Internet addiction based on personality characteristics of high school students in Kerman, Iran. Addict Health. 2011;3:85–91.PubMedPubMed CentralGoogle Scholar
  7. Senormancı O, Konkan R, Güçlü O, Senormancı G. Depression, loneliness, anger behaviours and interpersonal relationship styles in male patients admitted to internet addiction outpatient clinic in Turkey. Psychiatr Danub. 2014;26:39–45.PubMedGoogle Scholar
  8. Seo M, Kang HS, Yom YH. Internet addiction and interpersonal problems in Korean adolescents. Comput Inform Nurs. 2009;27:226–33.View ArticlePubMedGoogle Scholar
  9. Zhang MWB, Lim RBC, Lee C, Ho RCM. Prevalence of internet addiction in medical students: a meta-analysis. Acad Psychiatry. 2017. https://doi.org/10.1007/s40596-017-0794-1.Google Scholar
  10. Kaewpradub N, Kiatrungrit K, Hongsanguansri S, Pavasuthipaisit C. Association among internet usage, body image and eating behaviors of secondary school students. Shanghai Arch Psychiatry. 2017;29:208–17.PubMedPubMed CentralGoogle Scholar
  11. Boonvisudhi T, Kuladee S. Association between Internet addiction and depression in Thai medical students at Faculty of Medicine, Ramathibodi Hospital. PLoS ONE. 2017;12:e0174209.View ArticlePubMedPubMed CentralGoogle Scholar
  12. Jelenchick LA, Becker T, Moreno MA. Assessing the psychometric properties of the internet addiction test (IAT) in US college students. Psychiatry Res. 2012;196:296–301.View ArticlePubMedPubMed CentralGoogle Scholar
  13. Faraci P, Craparo G, Messina R, Severino S. internet addiction test (IAT): which is the best factorial solution? J Med Internet Res. 2013;9:e225.View ArticleGoogle Scholar
  14. Lai C-M, Mak K-K, Watanabe H, Ang RP, Pang JS, Ho RCM. Psychometric properties of the internet addiction test in Chinese adolescents. J Pediatr Psychol. 2013;38:794–807.View ArticlePubMedGoogle Scholar
  15. Boysan M, Kuss DJ, Barut Y, Ayköse N, Güleç M, Özdemir O. Psychometric properties of the Turkish version of the internet addiction test (IAT). Addict Behav. 2017;64:247–52.View ArticlePubMedGoogle Scholar
  16. Fernández-Villa T, Molina AJ, García-Martín M, Llorca J, Delgado-Rodríguez M, Martín V. Validation and psychometric analysis of the internet addiction test in Spanish among college students. BMC Public Health. 2015;15:953.View ArticlePubMedPubMed CentralGoogle Scholar
  17. Young K. Internet addiction: the emergence of a new clinical disorder. CyberPsychol Behav. 1998;1:237–44.View ArticleGoogle Scholar
  18. Lee K, Lee H-K, Gyeong H, Yu B, Song Y-M, Kim D. Reliability and validity of the Korean version of the internet addiction test among college students. J Korean Med Sci. 2013;28:763–8.View ArticlePubMedPubMed CentralGoogle Scholar
  19. Chang-Kook Y, Byeong-Moo C, Matthew B, Jeong-Hyeong L, Jin-Seok C. SCL-90-R and 16PF profiles of senior high school students with excessive internet use. Canadian J Psychiatry. 2005;50:407–14.View ArticleGoogle Scholar
  20. Barke A, Nyenhuis N, Kröner-Herwig B. The German version of the internet addiction test: a validation study. Cyberpsychol Behav Soc Netw. 2012;15:534–42.View ArticlePubMedGoogle Scholar
  21. Khazaal Y, Billieux J, Thorens G, Khan R, Louati Y, Scarlatti E, Theintz F, Lederrey J, Linden M, Zullino D. French validation of the internet addiction test. Cyberpsychol Behav. 2008;11:703–6.View ArticlePubMedGoogle Scholar
  22. Servidio R. Assessing the psychometric properties of the internet addiction test: a study on a sample of Italian university students. Comput Hum Behav. 2017;68:17–29.View ArticleGoogle Scholar
  23. Pontes HM, Patrão IM, Griffiths MD. Portuguese validation of the internet addiction test: an empirical study. J Behav Addict. 2014;3:107–14.View ArticlePubMedPubMed CentralGoogle Scholar
  24. Wartberg L, Petersen KU, Kammerl R, Rosenkranz M, Thomasius R. Psychometric validation of a German version of the compulsive internet use scale. Cyberpsychol Behav Soc Netw. 2014;17:99–103.View ArticlePubMedGoogle Scholar
  25. Chong Guan N, Mohammed Isa S, Hanim Hashim A, Pillai S, Kaur Harbajan Singh M. Validity of the Malay Version of the internet addiction test: a study on a group of medical students in Malaysia. 2012.Google Scholar
  26. Tran BX, Mai HT, Nguyen LH, Nguyen CT, Latkin CA, Zhang MWB, Ho RCM. Vietnamese validation of the short version of internet addiction test. Addict Behav Rep. 2017;6:45–50.View ArticleGoogle Scholar
  27. Korkeila J, Kaarlas S, Jääskeläinen M, Vahlberg T, Taiminen T. Attached to the web–harmful use of the Internet and its correlates. Eur Psychiatr. 2010;25:236–41.View ArticleGoogle Scholar
  28. Widyanto L, McMurran M. The psychometric properties of the internet addiction test. Cyberpsychol Behav. 2004;7:443–50.View ArticlePubMedGoogle Scholar
  29. Widyanto L, Griffiths MD, Brunsden V. A psychometric comparison of the internet addiction test, the internet-related problem scale, and self-diagnosis. Cyberpsychol Behav Soc Netw. 2010;14:141–9.View ArticlePubMedGoogle Scholar
  30. Wongpakaran T, Wongpakaran N. A short version of the revised ‘experience of close relationships questionnaire’: investigating non-clinical and clinical samples. Clin Pract Epidemiol Ment Health. 2012;8:36–42.View ArticlePubMedPubMed CentralGoogle Scholar
  31. Koyuncu T, Unsal A, Arslantas D. Assessment of internet addiction and loneliness in secondary and high school students. J Pak Med Assoc. 2014;64:998–1002.PubMedGoogle Scholar
  32. Li W, Zhang W, Xiao L, Nie J. The association of Internet addiction symptoms with impulsiveness, loneliness, novelty seeking and behavioral inhibition system among adults with attention-deficit/hyperactivity disorder (ADHD). Psychiatry Res. 2016;243:357–64.View ArticlePubMedGoogle Scholar
  33. Kline RB. Principles and practice of structural equation modeling. New York: Guilford; 1998.Google Scholar
  34. Li CH. Confirmatory factor analysis with ordinal data: comparing robust maximum likelihood and diagonally weighted least squares. Behav Res Methods. 2016;48:936–49.View ArticlePubMedGoogle Scholar
  35. Evaluating model fit. In: Hu L, Bentler PM editors. Structural equation modeling: concepts, issues and applications. California: Sage; 1995.Google Scholar
  36. Hu L, Bentler PM. Fit indices in covariance structure modeling: sensitivity to under parameterized model misspecification. Psychol Methods. 1998;3:424–53.View ArticleGoogle Scholar
  37. Hu L, Bentler PM. Cut off criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Model. 1999;6:1–55.View ArticleGoogle Scholar
  38. Muthen LK, Muthen BO. Mplus user’s guide, 7th edition. Los Angeles: Muthén & Muthén. 2015:1998–2015.Google Scholar
  39. Conti MA, Jardim AP, Hearst N, Cordás TA, Tavares H. Abreu CNd: avaliação da equivalência semântica e consistência interna de uma versão em português do internet addiction test (IAT). Arch Clin Psychiatr. 2012;39:106–10.Google Scholar
  40. Chang M, Law S. Factor structure for Young’s internet addiction test: a confirmatory study. Comput Human Behav. 2008;24:2597–619.View ArticleGoogle Scholar
  41. Tsimtsiou Z, Haidich AB, Kokkali S, Dardavesis T, Young KS, Arvanitidou M. Greek version of the internet addiction test: a validation study. Psychiatr Q. 2014;85:187–95.View ArticlePubMedGoogle Scholar
  42. Mohammadsalehi N, Mohammadbeigi A, Jadidi R, Anbari Z, Ghaderi E, Akbari M. Psychometric properties of the persian language version of yang internet addiction questionnaire: an explanatory factor analysis. Int J High Risk Behav Addict. 2015;4:e21560.View ArticlePubMedPubMed CentralGoogle Scholar
  43. Waqas A, Farooq F, Raza M, Javed ST, Khan S, Ghumman ME, Naveed S, Haddad M. Validation of the internet addiction test in students at a Pakistani Medical and Dental School. Psychiatr Q. 2017. https://doi.org/10.1007/s11126-017-9528-5.PubMedGoogle Scholar
  44. Nunnally JC, Bernstein IH. Psychometric theory. 3rd ed. New York: McGraw-Hill, Inc; 1994.Google Scholar
  45. Smith EV. Detecting and evaluating the impact of multidimensionality using item fit statistics and principal component analysis of residuals. J Appl Meas. 2002;3:205–31.PubMedGoogle Scholar
  46. Slocum-Gori SL, Zumbo BD: Assessing the unidimensionality of psychological scales: using multiple criteria from factor analysis. Soc Indic Res. 2011;102:443–61.Google Scholar
  47. Reeve BB, Hays RD, Bjorner JB, Cook KF, Crane PK, Teresi JA, Thissen D, Revicki DA, Weiss DJ, Hambleton RK, et al. Psychometric evaluation and calibration of health-related quality of life item banks: plans for the patient-reported outcomes measurement information system (PROMIS). Med Care. 2007;45:S22–31.View ArticlePubMedGoogle Scholar

Copyright

Advertisement