Erwin Handoko; Jekson Martiar Siahaan; Endy Juli Anto; Meldawati Meldawati; Muhammad Abdi Riyantono; Theresia Anggriani

doi:10.37104/ithj.v9i1.510

Erwin Handoko College of Medicine, Murni Teguh University, Indonesia
Jekson Martiar Siahaan College of Medicine, Murni Teguh University, Indonesia
Endy Juli Anto College of Medicine, Murni Teguh University, Indonesia
Meldawati Meldawati College of Medicine, Murni Teguh University, Indonesia
Muhammad Abdi Riyantono College of Medicine, Murni Teguh University, Indonesia
Theresia Anggriani College of Medicine, Murni Teguh University, Indonesia

DOI: https://doi.org/10.37104/ithj.v9i1.510

Keywords: Biostatistics, Statistical Analysis, Statistical Decision-Making, Health Research, HealthStat-11

Abstract

Selecting an appropriate statistical analysis method is an important step in producing health research findings that are valid, transparent, and clinically meaningful. However, the choice of statistical methods is often based on habit, familiarity with available software options, or oversimplified rules. For example, parametric tests are commonly used for “normally” distributed data and nonparametric tests for data that are considered “non-normal.” Distributional assumptions are important in selecting statistical analyses, but relying on these assumptions as the sole basis for choosing a statistical test is inadequate. The selection of a statistical test should instead begin with the research question and study design, followed by consideration of other factors, such as the measurement scale of the dependent variable and the sample size. We conducted a methodological review of guidance on statistical method selection in health research to translate these considerations into a practical sequence of decisions. Through searches of the PubMed, Scopus, and Web of Science databases, we identified 102 records. After removing 45 duplicates, 57 unique records were screened. Of these, 15 full-text articles were assessed for eligibility, and ultimately 14 articles were included in the review. In addition, the SAMPL, STROBE, and CONSORT 2025 guidelines were used as references to inform the reporting component. The synthesis yielded 11 interrelated principles, which we introduce as the HealthStat-11 Model. This framework was developed to guide health researchers in selecting statistical methods, interpreting their analyses, and reporting statistical findings. HealthStat-11 is not an automated algorithm for selecting statistical tests. Rather, it is intended to make explicit the reasoning underlying analytical choices and to identify points at which statistical consultation may be warranted.

Abstrak

Pemilihan metode analisis statistik yang tepat merupakan langkah yang penting untuk menghasilkan hasil penelitian kesehatan yang valid, transparan, dan bermakna secara klinis. Akan tetapi, pemilihan metode penelitian sering didasarkan atas kebiasaan, menu aplikasi yang dikenal, atau aturan-aturan yang disederhanakan. Sebagai contoh, penggunaan uji parametik untuk data yang terdistribusi normal dan nonparametrik untuk yang “tidak normal”. Asumsi distribusi cukup penting dalam pemilihan analisis statistik, tetapi menggunakan asumsi ini sebagai satu-satunya dasar pemilihan uji statistik adalah kurang tepat. Keputusan pemilihan uji statistik sebaiknya dimulai dari pertanyaan penelitian dan desain penelitian, lalu mempertimbangkan aspek lainnya seperti jenis skala pengukuran untuk variabel dependen, hingga besaran sampel. Kami melakukan tinjauan metodologis terhadap panduan-panduan dalam pemilihan metode statistik di bidang kesehatan untuk menerjemahkan keputusan-keputusan tersebut dalam urutan yang praktis. Dari penelusuran database PubMed, Scopus, dan Web of Science, kami mengidentifikasi 102 artikel penelitian; setelah mengeluarkan 45 artikel duplikat, tersaring 57 artikel yang berbeda. Dari artikel-artikel tersebut, 15 artikel dengan teks lengkap dianalisis kelayakannya, dan pada akhirnya, 14 artikel dimasukkan dalam kajian. Selain itu, pedoman-pedoman SAMPL, STROBE, dan CONSORT 2025 juga digunakan sebagai acuan dalam penyusunan laporan penelitian. Dari hasil analisis tersebut, dihasilkan 11 prinsip-prinsip yang saling berkaitan, yang kami perkenalkan sebagai Model Healthstat-11. Kerangka ini dikembangkan untuk dapat memandu peneliti-peneliti kesehatan dalam memilih metode statistik, menginterpretasikan, hingga melaporkan analisis statistik. ealthStat-11 bukan merupakan algoritma otomatis untuk memilih uji statistik. Sebaliknya, kerangka ini bertujuan untuk memperjelas dasar penalaran yang mendukung suatu pilihan analisis serta mengidentifikasi tahapan ketika konsultasi statistik mungkin diperlukan.

References

1. Kim J, Kim DH, Kwak SG. Comprehensive guidelines for appropriate statistical analysis methods in research. Korean J Anesthesiol. 2024;77(5).
2. Mishra P, Pandey C, Singh U, et al. Selection of appropriate statistical methods for data analysis. Ann Card Anaesth. 2019;22(3).
3. Ranganathan P. An introduction to statistics: Choosing the correct statistical test. Indian Journal of Critical Care Medicine. 2021;25(S2).
4. Najmi A, Sadasivam B, Ray A. How to choose and interpret a statistical test? An update for budding researchers. J Family Med Prim Care. 2021;10(8).
5. Arredondo Montero J. A Structured Guide to Univariate Test Selection Based on Normality, Variance Homogeneity, and Graphical Data Exploration. Journal of Surgical Research. 2026;318.
6. Poncet A, Courvoisier DS, Combescure C, et al. Normality and sample size do not matter for the selection of an appropriate statistical test for two-group comparisons. Methodology. 2016;12(2).
7. Vrbin CM. Parametric or nonparametric statistical tests: Considerations when choosing the most appropriate option for your data. Cytopathology. 2022;33(6).
8. Allen PJ, Dorozenko KP, Roberts D. Difficult decisions: A qualitative exploration of the statistical decision-making process from the perspectives of psychology students and academics. Front Psychol. 2016;7.
9. Hopewell S, Chan AW, Collins GS, et al. CONSORT 2025 statement: Updated guideline for reporting randomised trials. BMJ. 2025;389.
10. Lang TA, Altman DG. Basic statistical reporting for articles published in Biomedical Journals: The “Statistical analyses and methods in the published literature” or the SAMPL guidelines. Int. J. Nurs. Stud. 2015.
11. von Elm E, Altman DG, Egger M, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. International Journal of Surgery. 2014;12(12).
12. Barnett I, Torous J, Staples P, et al. Beyond smartphones and sensors: Choosing appropriate statistical methods for the analysis of longitudinal data. Journal of the American Medical Informatics Association. 2018.
13. Margarian A. Beyond P-Value-Obsession: When are Statistical Hypothesis Tests Required and Appropriate? German Journal of Agricultural Economics. 2022;71(4).
14. [Beath A, Jones MP. Guided by the research design: Choosing the right statistical test. Medical Journal of Australia. 2018;208(4).
15. Poythress JC, Lee MY, Young J. Planning and analyzing clinical trials with competing risks: Recommendations for choosing appropriate statistical methodology. Pharm. Stat. 2020.

PEMILIHAN METODE ANALISIS STATISTIK DALAM PENELITIAN KESEHATAN: PRINSIP DAN KERANGKA HEALTHSTAT-11

Abstract

References