EURASIAN JOURNAL OF

FAMILY MEDICINE

AVRASYA AİLE HEKİMLİĞİ DERGİSİ

Total Visitors: 141,405

 

Indexing/Abstracting

SCOPUS***ACCEPTED

Index Copernicus International

TR Index by TUBITAK ULAKBIM

Turkiye Citation Index

J-Gate

Turk Medline

InfoBase Index

Google Scholar



The Reliability and Validity Analysis of The Turkish Version of the Test to Assess the Psychological Dependence on Smoking



Mert Bardakci, Serdar Oztora, Hamdi Nezih Dagdeviren

Euras J Fam Med 2021;10(2):100-6. doi:10.33880/ejfm.2021100208





Original Research


ABSTRACT

Aim: Psychological component of cigarette dependence is as important as its physical component. In addition to psychoactive material in the cigarette, factors such as happiness, joy, accompaniment with another activity, feeling self-confident confidence and social status play an effective role in smoking dependence. This signifies the fact that smoking dependence has to be dealt with in a multidimensional way. Test to Assess the Psychological Dependence on Smoking (TAPDS) published by Ponciano-Rodríguez et al, is developed to assess the psychological dependence on tobacco. The aim of this study iis study aims to test the reliability and validity of the Turkish version of the Test to Assess the Psychological Dependence on Smoking in the Turkish population.

Methods: The Test to Assess the Psychological Dependence on Smoking (TAPDS) Turkish Version with 25 questions along with a sociodemographic form was handed out to volunteer participants. Data was were collected from a total number of 420 voluntary participants, who were randomly selected and current smokers. Confirmatory Factor Analysis was performed and Cronbach's α and McDonald's ω coefficients were used to determine the reliability of the scores of the scale.

Results: The participants consisted of 263 males and 157 females, with an average age of 31.07 years, ranging from 20 to 85. The factor loadings of the scale for all items are observed to be more than 0.30 and all the items are statistically significant. Cronbach's α and McDonald's ω values calculated for the total score are 0.93.


Conclusion: The Turkish version of TAPDS appears to be effective for assessing psychological dependence of smoking and it can be recommended as a sufficient and highly reliable and valid scale to be used.


Keywords: smoking, validity, reliability, psychological dependence


Introduction


The use of cigarettes/tobacco is one of the main causes of mortality and morbidity all over the world which can be prevented. According to data from WHO (World Health Organization), nearly 8 million people die of causes related with to smoking all over the world every year. More than 7 million of which are closely related to active smoking, and almost 1,2 million are those people that are passively exposed to cigarette smoke (1,2). Smoking is the main cause of chronic obstructive pulmonary disease (COPD), lung cancer, and atherosclerosis and is directly related to many fatal diseases. Besides, the use of tobacco leads to unhealthy life as well as loss of workforce and money (3,4).


Nicotine in the cigarette creates dependence since it changes the equilibrium of the two chemicals called dopamine and noradrenalin by activating the rewarding system of the brain (5). Physical dependence means compulsive use of nicotine, and nicotine leads to physical symptoms called “withdrawal syndrome” after consuming the cigarette (6,7). Psychological The psychological component of cigarette dependence is as important as its physical component. In addition to psychoactive material in the cigarette which causes addiction, it is considered that factors such as happiness, joy, accompaniment with another activity, feeling self-confident confidence and social status play an effective role in smoking dependence. This signifies the fact that smoking dependence is one which has to be dealt with in a multidimensional way (7,8).


Cigarette / /tobacco dependency is handled in detail within the chapter of Disorders of Drugs Use in the 5th edition of The Diagnostic and Statistical Manual of Mental Disorders (DSM-5). Within 12 months, diagnosis can be established based on the intolerance and withdrawal symptoms, considering whether or not it causes deterioration and trouble in the functionality of the person (9).

A great number of tests and scales have been formed and used to assess smoking dependence. Each one of these scales has different advantages and disadvantages in order to identify and analyze the parameters of physical and psychological dependence on smoking. Thus, the most suitable one could be preferred according to the purpose (7, 9).


Test to Assess the Psychological Dependence on Smoking (TAPDS) is a scale which that was developed and published by Ponciano-Rodríguez and his colleagues, to test psychological dependence on tobacco for the smokers in the Mexican population (7). This scale, which will be helpful both in clinical research and in the treatment of smoking dependence, is consisted of 25 questions. During the course of outlining the scale items, 100 smokers out of those who came to apply for a “Tobacco Cessation Clinic” in Mexico were asked to write down 5 words that come into mind in order of importance when the word “cigarette” is pronounced to them, and 23 words they named the most were identified. Additionally, letters called “farewell to cigarette”, which is part of tobacco cessation treatment, were read thoroughly. Keywords conveying emotional content in these letters were chosen by four psychologists who made contribution duringcontributed to the treatment. 40 words obtained were used to form The Test to Assess the Psychological Dependence on Smoking (TAPDS). The scale eventually turned into its final form of 25 items in total which has adequate information about the four subscales that include emotional modulator, image/self-rewarding, indifference,and social acceptance (7,9).


This study is aimed at testing the reliability and validity of the Turkish version of the Test to Assess the Psychological Dependence on Smoking (TAPDS) in the Turkish population.

Methods


In order tTo adapt TAPDS to Turkish culture and to make a validity and reliability analysis of the Turkish version of the scale, the original version was received from the original researcher Guadalupe Ponciano-Rodríguez, along with his permission. Later, it was translated into Turkish and re-translated into English by two English native speakers plus two English teachers and thus it was made sure that the translated one was suitable in meaning for use.


As of March 2021, the Test to Assess the Psychological Dependence on Smoking (TAPDS) Turkish Version (Figure 1) with 25 questions was handed out to volunteer participants. Data was were collected from a total number of 420 voluntary participants, who were randomly selected and current smokers.

Figure 1. The Turkish Version of Test to Assess the Psychological Dependence on Smoking (TAPDS)


Confirmatory Factor Analysis was used to show that the original structure of TAPDS is similar to Turkish culture. The items in the scale were structured using three gradations (Very frequently=3, Sometimes=2, Never=1). Therefore, the answers given to the items were identified categorically for confirmatory factor analysis, and WLSMV (Weighted Least Square Mean and Variance Adjusted Estimators) was used for that purpose. In the evaluation of model data compatibility, the CFI (Comparative Fit Index), TLI (Tucker Lewis index), NFI (Normed Fit Index) ve RMSEA (Root Mean Square Error of Approximation) index were was used. Statistics was were done using lavaan package of R 4.0.5 (R Core Team, 2018). Moreover, Cronbach's α and McDonald's ω coefficients were used to determine the reliability of the scores of the scale.

Results


Data was were collected from a total number of 420 participants consisted of 263 male and 157 female smokers for the TAPDS scale. Out of 420 participants, 237 were single, 171 were married and 12 divorced. The ages of the participants range from 20 through 85. The average age is 31,07.


TAPDS is composed of 25 questions and four subscales. In the course of making confirmatory factor analysis, model data compatibility was assessed by studying factor loadings and fit index. Factor loadings and the values of significance are given in Table 1.

Table 1. Factor loadings obtained as a result of confirmatory factor analysis


Item No
 
Emotional modulator Image/Self rewarding Indifference 
Social Acceptance
 
Z 
p

2
 
0.790
 
 
 
 
 
 
 
 
 
 


14
 
0.763
 
 
 
 
 
 
 
19.601


15 0.898
 
 
 
 
 
 
 
23.316
 
<0.05

16
 
0.869
 
 
 
 
 
 
 
22.762


17 
0.817
 
 
 
 
 
 
 
19.831


18
 
0.851
 
 
 
 
 
 
 
22.812


19
 
0.912
 
 
 
 
 
 
 
23.376


20
 
0.835
 
 
 
 
 
 
 
21.531

3 
   
 
 0.777 
 
   
 
   
 
   
 
   


 4 
 
   
 
 0.757 
 
   
 
   
 
 20.018 


 5 
 
   
 
 0.809 
 
   
 
   
 
 22.586 


 6 
 
   
 
 0.803 
 
   
 
   
 
 21.463 


 7 
 
   
 
 0.828 
 
   
 
   
 
 21.981 


 8 
 
   
 
 0.797 
 
   
 
   
 
 17.543 


 9 
 
   
 
 0.638 
 
   
 
   
 
 12.772 


 10 
 
   
 
 0.825 
   
 
   
 
 23.546 


 21 
 
   
 
   
 
 0.957 
 
   
 
   
 
   


 22 
 
   
 
   
 
 0.943 
 
   
 
 34.712 


 23 
 
   
 
   
 
 0.859 
 
   
 
 29.700 


 24 
 
   
 
   
 
 0.821 
 
   
 
 24.024 


 25 
 
   
 
   
 
 0.820 
 
   
 
 30.318 


 1 
 
   
 
   
 
   
 
 0.473 
 
   
 
   


 11 
 
   
 
   
 
   
 
 0.709 
 
 7.861 


 12 
 
   
 
   
 
   
 
 0.572 
 
 7.744 


 13 
 
   
 
   
 
   
 
 0.366 
 
 5.531 


According to Table 1, the factor loadings of the scale for all items are observed to be more than 0.30 and all the items are statistically significant (p<0,05). Fit index obtained as a result of confirmatory factor analysis is presented in Table 1. In addition, the measurement model obtained as a result of the analysis is given in Figure 2.

Figure 2. Measurement model of The Test to Assess the Psychological Dependence on Smoking (TAPDS) for Confirmatory Factor Analysis (Note that “M2-M14-M15-…” refer to “Item2-Item14-Item15-…”)

Examining Table 2, χ2/sd value is seen to be between 3 and 5. CFI (Comparative Fit Index) value is determined as 0.95, TLI (Tucker Lewis index) value as 0.94, and NFI (Normed Fit Index) as 0,93. With respect to RMSEA (Root Mean Square Error of Approximation) index , it is determined to be 0,086 for this model. Cronbach's α and McDonald's ω values for the total score are calculated as 0.93.

Table 2. Fit index results of confirmatory factor analysis of the test to assess the psychological dependence on smoking (TAPDS)


 
 
 χ2 
 
 χ2/sd 
 
 p 
 
 CFI 
 
 TLI 
 
 NFI 
 
 RMSEA 


 Scale 
 
 1111.067 
 
 4.13 
 
 0,00 
 
 0.95 
 
 0.94 
 
 0.93 
 
 0.086 


 Reference 
 
   
 
 χ2/sd≤3 
 
   
 
 ≥90 
 
 ≥90 
 
 ≥90 
 
 ≤0,080 



To determine the reliability of the scores of TAPDS, the reliability values of Cronbach's α and McDonald's ω were calculated. The results are shown in Table 3.

Table 3. Assessment of TAPDS scores’ reliability



































   
 
 Number of items 
 
 Cronbach's α 
 
 McDonald's ω 

Emotional modulator 
 
 8 
 
 0.90 
 
 0.90 


 Image/Self rewarding 
 
 8 
 
 0.86 
 
 0.86 


 Indifference 
 
 5 
 
 0.87 
 
 0.88 


 Social Acceptance 
 
 4 
 
 0.51 
 
 0.53 



As shown in Table 4, α and ω values for the emotional modulator subscale is 0.90, α and ω values for image/self-rewarding subscale is 0,86, for indifference α value is 0,87 and ω value is 0.88, for social acceptance α value is 0,51 and ω value is 0.53.

Table 4. Factor loadings obtained from the results of second second-level confirmatory factor analysis

Item No Emotional 
modulator
Image/
Self rewarding
Indifference Social Acceptance TAPDS Z p
2 0.790            
14 0.764         19.611 <0.05
15 0.898         23.329 <0.05
16 0.869         22.766 <0.05
17 0.817         19.847 <0.05
18 0.851         22.830 <0.05
19 0.912         23.394 <0.05
20 0.835         21.537 <0.05
3   0.778          
4   0.757       20.002 <0.05
5   0.809       22.538 <0.05
6   0.803       21.443 <0.05
7   0.828       21.956 <0.05
8   0.797       17.501 <0.05
9   0.637       12.712 <0.05
10   0.825       23.543 <0.05
21     0.957        
22     0.943     34.640 <0.05
23     0.859     29.709 <0.05
24     0.821     24.044 <0.05
25     0.820     30.345 <0.05
1       0.473      
11       0.711   7.757 <0.05
12       0.573   7.665 <0.05
13       0.367   5.497 <0.05
Emotional modulator   0.848    
Image/Self Self-rewarding   0.922 15.542 <0.05
Indifference     0.593 13.029 <0.05
Social Acceptance     0.979 7.584 <0.05

In Figure 2, correlation values between the subscales of TAPDS are seen to vary between 0.518 and 0.931. While doing second second-level confirmatory factor analysis, model data compatibility was evaluated by studying the factor loadings and fit index. Factor loadings and the values of significance are given in Table 4.

According to Table 4, the factor loadings of the scale for all items are observed to be more than 0.30 and all items are statistically significant (p<0,05). In addition, factor loadings related to the total TAPDS score, which was defined upper to the subscales, are determined to vary between 0.593 and 0.979. Fit index obtained as a result of second second-level confirmatory factor analysis are is given in Table 4. Besides, the measurement model drawn out of the analysis is shown in Figure 3.

Figure 3. Measurement model of The Test to Assess the Psychological Dependence on Smoking (TAPDS) for Second Level Confirmatory Factor Analysis (Note that “M2-M14-M15-…” refer to “Item2-Item14-Item15-…”)

Table 5 shows that, χ2/sd value is located between 3 and 5. CFI value is determined to be 0.95, TLI value 0.94, and NFI value is 0.93. In terms of RMSEA index evaluation, this index was seen to be 0.085. Cronbach's α and McDonald's ω values calculated for scale total score determined at second level is 0.93.

Table 5. Fit index results of confirmatory factor analysis of the test to assess the psychological dependence on smoking (TAPDS)


   
 
 χ2 
 
 χ2/sd 
 
 p 
 
 CFI 
 
 TLI 
 
 NFI 
 
 RMSEA 


 Scale 
 
 1093.508 
 
 4.04 
 
 0.00 
 
 0.95 
 
 0.94 
 
 0.93 
 
 0.085 


 Reference 
 
   
 
 χ2/sd≤3 
 
   
 
 ≥90 
 
 ≥90 
 
 ≥90 
 
 ≤0.080 


Discussion


In our study, factor loadings for all of the items are seen to be high and statistically significant (p<0,05). Thus, it can be said that the items do serve to measuring measure the structure in the factor they belong to. (10).

In our research, χ2/sd values are seen to be between 3 and 5. So it can be derived from this result that the model is compatible with the data at a medium level. With a CFI value of 0.95, TLI value of 0.94, and NFI value of 0.93, it can be inferred that the model is compatible with the data quite well, since all of this these values are over 0.90. The RMSEA index is 0.085, and it is fairly close to the accepted value of 0.080 for good model compatibility according to the RMSEA index. In a general view of all the compatibility indexes at hand, the four subscale model is seen to be compatible with the data. (11).


According to the results of our study, α and ω values for the emotional modulator subscale is 0.90, α and ω values for image/self-rewarding subscale is 0.86, for indifference α value is 0.87 and ω value is 0.88, for social acceptance α value is 0.51 and ω value is 0.53. Since the values under 0.50 are regarded as low reliability, values between 0.50 and 0.80 indicate medium reliability and those over 0.80 mean high reliability (12), it can be concluded that scores obtained from the emotional modulator, image/self-rewarding, and indifference subscales are highly reliable, and the scores from social acceptance subscale are reliable at a medium level (12).


TAPDS is evaluated in terms of its total score. For that reason, the total TAPDS score is defined over four subscales. The fact that correlation values between the subscales of the TAPDS are varying between 0.518 and 0.931 may imply that an upper upper-level structure exists above the one at the first level. While doing second second-level confirmatory factor analysis, model data compatibility was evaluated by studying the factor loadings and fit index.


Factor loadings for all the items are seen to be higher than 0.30 and all the items are statistically significant (p<0,05). As a result, it can be said that items do serve to measuring measure the structure that they belong to. In addition, factor loadings related to the total TAPDS score which was defined upper to the subscales are determined to be varying between 0.593 and 0.979 and these values are higher than 0.30. (10).


The total scale score of our study, with a Cronbach's α and McDonald's ω values of 0.93, is seen to be compatible with the internal reliability analysis results (Cronbach's α=0.84) of the original scale developed by Ponciano-Rodríguez et al (7).


Increase An increase in the number of scales on smoking dependence that can assess different criteria and that can be preferred accordingly will definitely provide more opportunity to the health care professionals in determining smoking dependence and also in their treatment choice.


Conclusion


According to our study results, factor loadings of all the 25 items of the Turkish version of TAPDS were higher than 0.30 and all the items were statistically significant. Cronbach's α and McDonald's ω values were calculated as 0.93 for the total score of the scale and therefore the total score obtained from TAPDS was found to be reliable. The Turkish version of TAPDS appears to be effective for assessing psychological dependence of smoking and it can be recommended as a sufficient and highly reliable and valid scale to be used.


Acknowledgment


Special thanks to Tunçbilek Bardakcı, Semra Bardakcı, Esra Usta, Zeynep Usta and Gökhan Karakoç for their valuable contribution in this study.


References

  1. OECD/World Health Organization. Health at a Glance: Asia/Pacific 2020: measuring progress towards universal health coverage,. Paris: OECD Publishing, Paris. 2020. doi: 10.1787/26b007cd-en.
  2. Türkiye Halk Sağlığı Kurumu [Internet]. TC Sağlık Bakanlığı Küresel yetişkin tütün araştırması Türkiye 2012 [cited 2021 Apr 15]. Available from: https://havanikoru.saglik.gov.tr/dosya/dokumanlar/yayinlar/KYTA-2012-TR-25-07-2014.pdf.Sağlık Bakanlığı Yayın. 2014(948).
  3. Karlıkaya C, Öztuna F, Solak ZA, Özkan M, Örsel O. Tütün kontrolü. Toraks Ddergisi. 2006;7(1):51-64.

  4. Iribarren C, Tekawa IS, Sidney S, Friedman GD. Effect of cigar smoking on the risk of cardiovascular disease, chronic obstructive pulmonary disease, and cancer in men. New England Journal of Medicine. 1999;340(23):1773-80. doi: 10.1056/NEJM199906103402301.
  5. Acar S, Kaylı DŞ, Yararbaş G. Sigara kullanan, sigara bırakma tedavisi alan ve sigara kullanmayan bireylerin psikolojik dayanıklılık ve stresle başa çıkma tutumları bakımından karşılaştırılması. The Turkish Journal on Addictions. 2019;6:539-66. doi:10.15805/addicta.2019.6.3.0029.
  6. Türk Toraks Derneği Tütün Kontrolü Çalışma Grubu [Internet]. Tütün Kontolü Çalışma Grubu Sigara bırakma tanı ve tedavi uzlaşı raporu [cited 2021 Apr 10]. Available from: https://toraks.org.tr/site/community/library/1655. Ankara; 2014.
  7. Ponciano-Rodríguez G, Valerio-Gutiérrez R, Pliego-Rosas C, Córdova-Alcaráz A. Development and evaluation of a new self-rating test to assess the psychological dependence on smoking (TAPDS). J Addict Med Ther. 2015;3(2):1016.
  8. Işiktaş S, Karafistan M, Dilan A, Yilmaz AS. Sigara içen ve içmeyen bireylerin stresle başa çıkma tutumları açısından karşılaştırılması. Kıbrıs Türk Psikiyatri ve Psikoloji Dergisi. 2019;1(2):102-7. doi:10.35365/ctjpp.1.12.
  9. Sağlam L. Nikotin Bağımlılığının klinik değerlendirilmesi. Güncel Göğüs Hastalıkları Serisi. 2017;4(1):78-89. doi: 10.5152/gghs.2016.011.
  10. Büyüköztürk Ş. Faktör analizi: Temel kavramlar ve ölçek geliştirmede kullanımı. Kuram ve Uygulamada Eğitim Yönetimi,2002;32(32):470-483.
  11. İlhan M. ve Çetin B. LISREL ve AMOS programları kullanılarak gerçekleştirilen yapısal eşitlik modeli (YEM) analizlerine ilişkin sonuçların karşılaştırılması. Eğitimde ve Psikolojide Ölçme ve Değerlendirme Dergisi. 2014;5(2):26-42. doi: http://dx.doi.org/10.21031/epod.31126.
  12. Salvucci S, Walter E, Conley V, Fink S, Saba M. Measurement error studies at the National Center for Education Statistics. July 1997. Report no: NCES-97-464.

How to cite: Bardakci M, Oztora S, Dagdeviren HN. The reliability and validity analysis of the Turkish version of the Test To Assess The Psychological Dependence On Smoking. Euras J Fam Med 2021;10(2):100-6. doi:10.33880/ejfm.2021100208.


Download Full Text Add to Favorites