A Web-Based Guide in Choosing the Appropriate Test Statistics for Univariate Models

Mehmet Ali Sungur; Handan Ankaralı; Şengül Cangür

doi:10.21673/anadoluklin.284434

Research Article

A Web-Based Guide in Choosing the Appropriate Test Statistics for Univariate Models

Year 2017, , 24 - 29, 30.12.2016

Mehmet Ali Sungur , Handan Ankaralı , Şengül Cangür

https://doi.org/10.21673/anadoluklin.284434

Abstract

Aim: The rate of statistical software use among researchers today is very high. In this study we
aimed to develop a web-based decision tree at the least possible confusion level that can provide
guidance in choosing the appropriate test statistics for researchers who want to perform
the basic statistical analyses in univariate models on their own.

Materials and Methods: The decision tree developed as a guide in determining the appropriate
test statistics in the statistical analysis of simple or multiple univariate models was created by
using PHP and JavaScript languages. Technical information on the concepts and tests was
provided in pop-ups for the guide user’s convenience.

Results: The questions posed within this decision tree that is quite easy to reach and use help
reduce possible appropriate methods by fixing the research’s aim, design and data type and
lower the researcher’s possibility to make mistakes due to lack of knowledge or misguidance.
In addition, the information provided on the existence and nature of more advanced statistical
methods emphasizes for the researcher the importance of consulting with an expert during the
statistical analysis of a study’s results.

Discussion and Conclusion: Choosing the right test is crucial for the accuracy and reliability of findings to be obtained from
a statistical analysis. What is essential is the ability to choose the right test for statistical analysis, not the ability to perform
a statistical test or use statistical software. In addition to emphasizing the importance of consulting with an expert, a decision
tree as developed in this study will contribute to lowering the possibility of making mistakes of researchers who want to
perform statistical analyses on their own, at least while their choice of method.

Keywords

choosing appropriate test statistics, statistical analysis, univariate models.

References

1. Ankaralı H, Cangür Ş, Sungur MA. Formülsüz Biyoistatistik.İstanbul: BETİM; 2015.
2. Jaykaran D. How to select appropriate statistical test. J Pharm Negative Results. 2010;1(2):61–3.
3. Nayak BK, Hazra A. How to choose the right statistical test? Indian J Ophthalmol. 2011;59(2):85–6.
4. Johnson LR, Karunakaran UD. How to choose the appropriate statistical test using the free program statistics Open For All (SOFA). Annals of Community Health.2014;2(2):54–62.
5. Marusteri M, Bacarea V. Comparing groups for statistical differences: how to choose the right statistical test?Biochemia Medica. 2010;20(1):15–32.
6. Suner A, Karakülah G, Koşaner Ö, Dicle O. StatXFinder: A web-based self-directed tool that provides appropriate statistical test selection for biomedical researchers in their scientific studies. SpringerPlus. 2015;4:633.
7. Saltikov JB, Whittaker WJ. Selecting the most appropriate inferential statistical test for your quantitative research study. J Clin Nurs. 2013;23:1520–31.
8. Watt JH, van den Berg SA. Selecting Statistical Tests. Research Methods for Communication Science, 2. ed. Albany: Rensselaer Polytechnic Institute; 2002:302–29.
9. Parab S, Bhalerao S. Choosing statistical test. Int J Ayurveda Res. 2010;1(3):187–91.
10. Andrews FM, Klem L, Davidson TN, O’Malley PM, Rodgers WL. A Guide for Selecting Statistical Techniques for Analyzing Social Science Data, 2. ed. ABD: University of Michigan; 1981.
11. Naramore E, Gerner J, Scouarnec YL, Stolz J, Glass MK. Beginning PHP5, Apache, and MySQL web development. Indianapolis: Wiley; 2005:1–660.
12. Yank K. Introducing PHP. Build Your Own Database Driven Web Site Using PHP & MySQL, 4th ed. Victoria, AU: SitePoint Pty Ltd; 2009:73–113.
13. Rochkind M. Platforms and Tools. Expert PHP and MySQL. New York: Apress; 2013:45–81.
14. Suehring S, Converse T, ParkJ. PHP 6 and MySQL 6 Bible. Indianapolis: Wiley; 2009:1–793.
15. Alan MA, Gülmez M. The comparison of some database administration languages on the Internet. C.Ü. İktisadi
16. Stefanov S. JavaScript for PHP Developers. Sebastopol: O’Reilly; 2013:1–136.
17. Zakas NC. Professional JavaScript for Web Developers, 3. ed. Indianapolis: Wiley; 2012:1–856.
18. Morrison M. The Interactive Web: Reacting to the Virtual World. Head First JavaScript. Sebastopol: O’Reilly; 2008:1–31.
19. McFarland DS. Writing Your First JavaScript Program. JavaScript & jQuery, The Missing Manual, 2. ed. Sebastopol: O’Reilly; 2012:21–40.
Huberty CJ, Morris JD.,Multivariate analysis versus multiple univariate analyses.,302-8,1989,105
Day S.,Dictionary for Clinical Trials,,Wiltshire: Wiley; 2007:1–249.,2. ed.
Daniel WW, Cross CL.,Biostatistics: A Foundation for Analysis in the Health Sciences,,1–960,Wiley,10. ed. ABD:,2014

Tek Değişkenli Modellerde Uygun Test İstatistiğinin Seçilmesi İçin Web Tabanlı Bir Kılavuz

Year 2017, , 24 - 29, 30.12.2016

Mehmet Ali Sungur , Handan Ankaralı , Şengül Cangür

https://doi.org/10.21673/anadoluklin.284434

Abstract

Amaç: Araştırmacıların günümüzde istatistiksel paket program kullanma oranı çok yüksektir.
Bu çalışmada tek değişkenli modellerde temel düzeydeki istatistiksel analizleri yapmak isteyen
araştırmacılara uygun test istatistiğinin seçimi sırasında kılavuzluk edebilecek, mümkün olan en
az karmaşıklık düzeyinde web tabanlı bir karar ağacı geliştirmek amaçlanmıştır.

Gereç ve Yöntemler: Tek değişkenli basit veya çoklu modellerin istatistiksel analizinde hangi test
istatistiğinin uygun olduğunun belirlenişine yönelik bir kılavuz olarak geliştirilen karar ağacı, PHP
ve JavaScript dilleri yardımıyla oluşturulmuştur. Kılavuzun kolayca kullanılabilmesi için kavramlar
ve testlere ilişkin teknik bilgiler açılır pencereler içinde verilmiştir.

Bulgular: Ulaşması ve kullanması oldukça kolay olan karar ağacında araştırmacılara sorulan sorularla
çalışmanın amacı, tasarımı ve veri tipi saptanarak uygun yöntemler genelden özele doğru
indirgenmekte ve araştırmacının bilgi eksikliğinden ya da yanlış yönlendirmeden kaynaklanan
hata yapma olasılığı azaltılmaktadır. Ayrıca, ileri istatistiksel yöntemlerin varlığı ve özellikleri konusunda
araştırmacılar bilgilendirilerek uzman yardımı almanın önemi de vurgulanmıştır.

Tartışma ve Sonuç: İstatistiksel değerlendirme sonucunda elde edilecek bulguların doğruluğu
ve güvenilirliği açısından doğru testin seçilmesi oldukça önemlidir. En önemlisi paket program
kullanabilmek ya da bir testi uygulayabilmek değil, istatistiksel analiz için doğru testi seçebilmektir.
Geliştirilen karar ağacı, uzman yardımı almanın önemini vurgulamakla birlikte, temel dü-
zeydeki istatistiksel değerlendirmeleri kendi başına yapmak isteyen araştırmacıların en azından
yöntem seçimi sırasında hata yapma olasılığının azalmasına katkı sağlayacaktır

Keywords

uygun test istatistiği seçimi, istatistiksel analiz, tek değişkenli modeller

References

1. Ankaralı H, Cangür Ş, Sungur MA. Formülsüz Biyoistatistik.İstanbul: BETİM; 2015.
2. Jaykaran D. How to select appropriate statistical test. J Pharm Negative Results. 2010;1(2):61–3.
3. Nayak BK, Hazra A. How to choose the right statistical test? Indian J Ophthalmol. 2011;59(2):85–6.
4. Johnson LR, Karunakaran UD. How to choose the appropriate statistical test using the free program statistics Open For All (SOFA). Annals of Community Health.2014;2(2):54–62.
5. Marusteri M, Bacarea V. Comparing groups for statistical differences: how to choose the right statistical test?Biochemia Medica. 2010;20(1):15–32.
6. Suner A, Karakülah G, Koşaner Ö, Dicle O. StatXFinder: A web-based self-directed tool that provides appropriate statistical test selection for biomedical researchers in their scientific studies. SpringerPlus. 2015;4:633.
7. Saltikov JB, Whittaker WJ. Selecting the most appropriate inferential statistical test for your quantitative research study. J Clin Nurs. 2013;23:1520–31.
8. Watt JH, van den Berg SA. Selecting Statistical Tests. Research Methods for Communication Science, 2. ed. Albany: Rensselaer Polytechnic Institute; 2002:302–29.
9. Parab S, Bhalerao S. Choosing statistical test. Int J Ayurveda Res. 2010;1(3):187–91.
10. Andrews FM, Klem L, Davidson TN, O’Malley PM, Rodgers WL. A Guide for Selecting Statistical Techniques for Analyzing Social Science Data, 2. ed. ABD: University of Michigan; 1981.
11. Naramore E, Gerner J, Scouarnec YL, Stolz J, Glass MK. Beginning PHP5, Apache, and MySQL web development. Indianapolis: Wiley; 2005:1–660.
12. Yank K. Introducing PHP. Build Your Own Database Driven Web Site Using PHP & MySQL, 4th ed. Victoria, AU: SitePoint Pty Ltd; 2009:73–113.
13. Rochkind M. Platforms and Tools. Expert PHP and MySQL. New York: Apress; 2013:45–81.
14. Suehring S, Converse T, ParkJ. PHP 6 and MySQL 6 Bible. Indianapolis: Wiley; 2009:1–793.
15. Alan MA, Gülmez M. The comparison of some database administration languages on the Internet. C.Ü. İktisadi
16. Stefanov S. JavaScript for PHP Developers. Sebastopol: O’Reilly; 2013:1–136.
17. Zakas NC. Professional JavaScript for Web Developers, 3. ed. Indianapolis: Wiley; 2012:1–856.
18. Morrison M. The Interactive Web: Reacting to the Virtual World. Head First JavaScript. Sebastopol: O’Reilly; 2008:1–31.
19. McFarland DS. Writing Your First JavaScript Program. JavaScript & jQuery, The Missing Manual, 2. ed. Sebastopol: O’Reilly; 2012:21–40.
Huberty CJ, Morris JD.,Multivariate analysis versus multiple univariate analyses.,302-8,1989,105
Day S.,Dictionary for Clinical Trials,,Wiltshire: Wiley; 2007:1–249.,2. ed.
Daniel WW, Cross CL.,Biostatistics: A Foundation for Analysis in the Health Sciences,,1–960,Wiley,10. ed. ABD:,2014

There are 22 citations in total.

Details

Subjects	Health Care Administration
Journal Section	ORIGINAL ARTICLE
Authors	Mehmet Ali Sungur Handan Ankaralı Şengül Cangür
Publication Date	December 30, 2016
Acceptance Date	August 16, 2016
Published in Issue	Year 2017

Cite

Vancouver	Sungur MA, Ankaralı H, Cangür Ş. A Web-Based Guide in Choosing the Appropriate Test Statistics for Univariate Models. Anadolu Klin. 2016;22(1):24-9.