Deneysel Araştırmalarda İstatistiksel Analiz
Özet
Deneysel sinirbilimde elde edilen davranışsal ve elektrofizyolojik ölçümler; çoğu zaman çok düzeyli karmaşık bir yapıya sahip olup istatistiksel yöntemin doğru seçilmesini oldukça önemli hale getirir. Bu bölümde, araştırma sorusunun türünden (fark/ilişki) başlayarak değişkenlerin tanımlanması, grupların bağımlı/bağımsız yapısının belirlenmesi ve veri tipinin sınıflandırılması üzerinden ilerleyen sistematik bir analiz akışı sunulmaktadır. Nicel verilerde parametrik testlerin geçerliliği açısından normallik varsayımı; histogram ve özellikle Q–Q grafiği gibi görsel araçlar ile Shapiro–Wilk, Kolmogorov–Smirnov/Lilliefors ve Anderson–Darling gibi analitik testler birlikte değerlendirilerek ele alınmaktadır. Örneklem büyüklüğünün normallik testlerinin duyarlılığı üzerindeki etkisi vurgulanarak, kararın tek bir p-değerine indirgenmemesi gerektiği belirtilmektedir. Varsayımlar makul ölçüde sağlandığında t-testi, ANOVA ve Pearson korelasyonu gibi parametrik yöntemler; ihlal durumlarında ise Mann–Whitney U, Wilcoxon, Kruskal–Wallis, Friedman ve Spearman korelasyonu gibi parametrik olmayan alternatifler önerilmektedir. Son olarak, çoklu karşılaştırmalarda post hoc yaklaşımların rolü özetlenmiştir.
Behavioral, imaging, and electrophysiological measurements obtained in experimental neuroscience often exhibit a complex, multilevel structure, making the appropriate choice of statistical method critical. This chapter presents a systematic analysis workflow that begins with the type of research question (difference vs. association) and proceeds through defining the variables, determining whether the groups/observations are independent or dependent, and classifying the data type. For quantitative data, the normality assumption—central to the validity of parametric tests—is addressed by jointly considering visual tools such as histograms and especially Q–Q plots, together with analytical tests including Shapiro–Wilk, Kolmogorov–Smirnov/Lilliefors, and Anderson–Darling. Emphasizing the effect of sample size on the sensitivity of normality tests, the chapter notes that decisions should not be reduced to a single p-value. When assumptions are reasonably satisfied, parametric methods such as the t-test, ANOVA, and Pearson correlation are recommended; when violations are present, nonparametric alternatives such as the Mann–Whitney U test, Wilcoxon test, Kruskal–Wallis test, Friedman test, and Spearman correlation are suggested. Finally, the role of post hoc approaches in multiple comparisons is summarized.
Referanslar
Truong, Y. K., & Lewis, M. M. (2016). Statistical techniques for neuroscientists. CRC Press.
Aarts, E., Dolan, C., Verhage, M., & Sluis, S. (2015). Multilevel analysis quantifies variation in the experimental effect while optimizing power and preventing false positives. BMC Neuroscience, 16(1). https://doi.org/10.1186/s12868-015-0228-5
Galbraith, S., Daniel, J., & Vissel, B. (2010). A study of clustered data and approaches to its analysis. Journal of Neuroscience, 30(32), 10601-10608. https://doi.org/10.1523/jneurosci.0362-10.2010
Ghasemi, A., & Zahediasl, S. (2012). Normality tests for statistical analysis: a guide for non-statisticians. International journal of endocrinology and metabolism, 10(2), 486.
Field, A. (2024). Discovering statistics using IBM SPSS statistics. Sage publications limited.
Lilliefors, H. W. (1967). On the Kolmogorov–Smirnov test for normality with mean and variance unknown. Journal of the American Statistical Association, 62(318), 399–402.
Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3–4), 591–611.
Jarque, C. M., & Bera, A. K. (1987). A test for normality of observations and regression residuals. International Statistical Review, 55(2), 163–172.
Stephens, M. A. (1974). EDF Statistics for Goodness of Fit and Some Comparisons, Journal of the American Statistical Association, 69, pp. 730-737.
Sheskin, D. J. (2011). Handbook of Parametric and Nonparametric Statistical Procedures (5th ed.). CRC Press.
Kurtz, A. K., & Mayo, S. T. (2013). Instructor’s manual for statistical methods in education and psychology. Springer Science & Business Media.
Arslan, K. (2024). Spss ile Adım Adım Bağımsız Örneklem T Testi Uygulama ve Apa Formatında Raporlama. F. Akça (Ed.), Eğitim Bilimleri Alanında Araştırmalar ve Değerlendirmeler (s. 125-138). Ankara: Gece Kitaplığı.
Ross, A., & Willson, V. L. (2018). Basic and advanced statistical tests: Writing results sections and creating tables and figures. Springer.
Ross, A., & Willson, V. L. (2017). One-way anova. In Basic and advanced statistical tests: Writing results sections and creating tables and figures (pp. 21-24). Rotterdam: SensePublishers.
Park, E., Cho, M., & Ki, C. S. (2009). Correct use of repeated measures analysis of variance. Korean J Lab Med, 29(1), 1-9.
Sedgwick, P. (2012). Pearson’s correlation coefficient. Bmj, 345.
Levene, H. (1960). Robust tests for equality of variances. In I. Olkin (Ed.), Contributions to Probability and Statistics: Essays in Honor of Harold Hotelling (pp. 278–292). Stanford University Press.
Greenhouse, S. W., Geisser, S. (1959). On methods in the analysis of profile data. Psychometrika, 24(2), 95–112.
Huynh, H., & Feldt, L. S. (1976). Estimation of the Box correction for degrees of freedom from sample data in randomized block and split-plot designs. Journal of Educational Statistics, 1(1), 69–82.
Sedgwick, P. (2014). Spearman’s rank correlation coefficient. Bmj, 349.
Chicco, D., Sichenze, A., & Jurman, G. (2025). A simple guide to the use of Student’s t-test, Mann-Whitney U test, Chi-squared test, and Kruskal-Wallis test in biostatistics. BioData Mining, 18(1), 56.
Orhunbilge, N., Örnekleme Yöntemleri ve Hipotez Testleri (2000). İstanbul Üniversitesi İşletme Fakültesi, İşletme İktisadı Yayın No: 8, s.182.
Benjamini, Y., Hochberg, Y. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B, 57(1), 289–300.
