Department of Health Policy and Management, College of Health Science, andDepartment of Public Health Science, Graduate School, Korea University, Seoul,Korea.
Correspondence to Hae-Young Kim, DDS, PhD. Professor, Department of Health
Policy and Management, Korea University College of Health Science, and
Department of Public Health Science, Korea University Graduate School, 145
Anam-ro, Seongbuk-gu, Seoul 02841, Korea. kimhaey@korea.ac.kr
This is an Open Access article distributed under the terms of the
Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which
permits unrestricted non-commercial use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The t-test is frequently used in comparing 2 group means. The compared
groups may be independent to each other such as men and women. Otherwise, compared data
are correlated in a case such as comparison of blood pressure levels from the same
person before and after medication (Figure 1). In
this section we will focus on independent t-test only. There are 2
kinds of independent t-test depending on whether 2 group variances can
be assumed equal or not. The t-test is based on the inference using
t-distribution.
The t-distribution was invented in 1908 by William Sealy Gosset, who
was working for the Guinness brewery in Dublin, Ireland. As the Guinness brewery did
not permit their employee's publishing the research results related to their work,
Gosset published his findings by a pseudonym, “Student.” Therefore,
the distribution he suggested was called as Student's
t-distribution. The t-distribution is a
distribution similar to the standard normal distribution,
z-distribution, but has lower peak and higher tail compared to it
(Figure 2).
Figure 2
The t-distribution with various degrees of freedom (df)
compared to z-distribution.
According to the sampling theory, when samples are drawn from a normal-distributed
population, the distribution of sample means is expected to be a normal
distribution. When we know the variance of population, σ2, we can
define the distribution of sample means as a normal distribution and adopt
z-distribution in statistical inference. However, in reality,
we generally never know σ2, we use sample variance,
s2, instead. Although the
s2 is the best estimator for σ2,
the degree of accuracy of s2 depends on the sample size.
When the sample size is large enough (e.g., n =
300), we expect that the sample variance would be very similar to the population
variance. However, when sample size is small, such as n = 10, we
could guess that the accuracy of sample variance may be not that high. The
t-distribution reflects this difference of uncertainty
according to sample size. Therefore the shape of t-distribution
changes by the degree of freedom (df), which is sample size minus one (n − 1)
when one sample mean is tested.
The t-distribution appears to be a family of distribution of which
shape varies according to its df (Figure 2).
When df is smaller, the t-distribution has lower peak and higher
tail compared to those with higher df. The shape of t-distribution
approaches to z-distribution as df increases. When df gets large
enough, e.g., n = 300,
t-distribution is almost identical with
z-distribution. For the inferences of means using small samples, it
is necessary to apply t-distribution, while similar inference can
be obtain by either t-distribution or
z-distribution for a case with a large sample. For inference of 2
means, we generally use t-test based on
t-distribution regardless of the sizes of sample because it is
always safe, not only for a test with small df but also for that with large df.
INDEPENDENT SAMPLES T-TEST
To adopt z- or t-distribution for inference using
small samples, a basic assumption is that the distribution of population is not
significantly different from normal distribution. As seen in Appendix 1, the normality assumption needs to be tested in
advance. If normality assumption cannot be met and we have a small sample
(n < 25), then we are not permitted to use
‘parametric’ t-test. Instead, a non-parametric
analysis such as Mann-Whitney U test should be selected.
For comparison of 2 independent group means, we can use a
z-statistic to test the hypothesis of equal population means only
if we know the population variances of 2 groups, σ21 and σ22, as follows;
(Eq. 1)
Z=−X1−−X2√σ21n1+σ22n2
where X̄1 and
X̄2, σ21 and σ22, and n1 and
n2 are sample means, population variances, and the
sizes of 2 groups.
Again, as we never know the population variances, we need to use sample variances as
their estimates. There are 2 methods whether 2 population variances could be assumed
equal or not. Under assumption of equal variances, the t-test
devised by Gosset in 1908, Student's t-test, can be applied. The
other version is Welch's t-test introduced in 1947, for the cases
where the assumption of equal variances cannot be accepted because quite a big
difference is observed between 2 sample variances.
1. Student's t-test
In Student's t-test, the population variances are assumed equal.
Therefore, we need only one common variance estimate for 2 groups. The common
variance estimate is calculated as a pooled variance, a weighted average of 2
sample variances as follows;
(Eq. 2)
s2p=(n1−1)(n1−1)+(n2−1)s21+(n2−1)(n1−1)+(n2−1)s22
where s21 and s22 are sample variances.
The resulting t-test statistic is a form that both the
population variances, σ21 and σ21, are exchanged with a common variance
estimate, s2p. The df is given as
n1 + n2 − 2
for the t-test statistic.
In Appendix 1, ‘(E-1) Leven's test
for equality of variances’ shows that the null hypothesis of equal
variances was accepted by the high p value, 0.334 (under
heading of Sig.). In ‘(E-2) t-test for equality of means
t-values’, the upper line shows the result of
Student's t-test. The t-value and df are shown
−3.357 and 18. We can get the same figures using the formulas Eq. 2 and
Eq. 3, and descriptive statistics in Table
1, as follows.
Table 1
Descriptive statistics and result of the Student's
t-test
The result of calculation is a little different from that by SPSS (IBM Corp.,
Armonk, NY, USA) of Appendix 1, maybe
because of rounding errors.
2. Welch's t-test
Actually there are a lot of cases where the equal variance cannot be assumed.
Even if it is unlikely to assume equal variances, we still compare 2 independent
group means by performing the Welch's t-test. Welch's
t-test is more reliable when the 2 samples have unequal
variances and/or unequal sample sizes. We need to maintain the assumption of
normality.
Because the population variances are not equal, we have to estimate them
separately by 2 sample variances, s21 and s22. As the result, the form of
t-test statistic is given as follows;
(Eq. 4)
t=−X1−−X2√s21n1+s22n2~tν
where ν is Satterthwaite degrees of freedom.
(Eq. 5)
ν=s21n1+s22n2(s21n1)2/(n1−1)+(s22n2)2/(n2−1)
In Appendix 1, ‘(E-1) Leven's test
for equality of variances’ shows an equal variance can be successfully
assumed (p = 0.334). Therefore, the Welch's
t-test is inappropriate for this data. Only for the purpose of
exercise, we can try to interpret the results of Welch's t-test
shown in the lower line in ‘(E-2) t-test for equality of
means t-values’. The t-value and df are
shown as −3.357 and 16.875.
We've confirmed nearly same results by calculation using the formula and by SPSS
software.
The t-test is one of frequently used analysis methods for
comparing 2 group means. However, sometimes we forget the underlying assumptions
such as normality assumption or miss the meaning of equal variance assumption.
Especially when we have a small sample, we need to check normality assumption
first and make a decision between the parametric t-test and the
nonparametric Mann-Whitney U test. Also, we need to assess the
assumption of equal variances and select either Student's
t-test or Welch's t-test.
Appendix
Appendix 1
Procedure of t-test analysis using IBM SPSS
The procedure of t-test analysis using IBM SPSS Statistics
for Windows Version 23.0 (IBM Corp., Armonk, NY, USA) is as follows.
Examining Gender Differences in Aggression as a Predictor of Anxiety, Depression, and Suicide in a Cross‐Sectional French Sample Sylvia Martin Health Science Reports.2025;[Epub] CrossRef
Structural and topological analysis of thiosemicarbazone-based metal complexes: computational and experimental study of bacterial biofilm inhibition and antioxidant activity Doaa S. El‑Sayed, Shaymaa S. Hassan, Liblab S. Jassim, Ali Abdullah Issa, Firas AL-Oqaili, Mustafa k. Albayaty, Buthenia A. Hasoon, Majid S. Jabir, Khetam H. Rasool, Hemmat A. Elbadawy BMC Chemistry.2025;[Epub] CrossRef
Comparison of the effects of video conference and video-based home exercise on physical performance and body composition in older adult individuals Özgün Elmas, Mustafa Cemali, Ayşe Livanelioğlu Medicine.2024; 103(44): e40329. CrossRef
Deep Residual Learning-Based Classification with Identification of Incorrect Predictions and Quantification of Cellularity and Nuclear Morphological Features in Digital Pathological Images of Common Astrocytic Tumors Yen-Chang Chen, Shinn-Zong Lin, Jia-Ru Wu, Wei-Hsiang Yu, Horng-Jyh Harn, Wen-Chiuan Tsai, Ching-Ann Liu, Ken-Leiang Kuo, Chao-Yuan Yeh, Sheng-Tzung Tsai Cancers.2024; 16(13): 2449. CrossRef
The Impact of Stress First Aid on Perceived Stress Levels of New Graduate Nurses John R. Balcuk Nursing Economic$.2024; 42(4): 191. CrossRef
CloudBrain-MRS: An intelligent cloud computing platform for in vivo magnetic resonance spectroscopy preprocessing, quantification, and analysis Xiaodie Chen, Jiayu Li, Dicheng Chen, Yirong Zhou, Zhangren Tu, Meijin Lin, Taishan Kang, Jianzhong Lin, Tao Gong, Liuhong Zhu, Jianjun Zhou, Ou-yang Lin, Jiefeng Guo, Jiyang Dong, Di Guo, Xiaobo Qu Journal of Magnetic Resonance.2024; 358: 107601. CrossRef
Bactericidal activities and biochemical features of 16 antimicrobial peptides against bovine-mastitis causative pathogens Hye-sun Cho, Dohun Kim, Hyoim Jeon, Prathap Somasundaram, Nagasundarapandian Soundrarajan, Chankyu Park Veterinary Research.2024;[Epub] CrossRef
Expressions of Interleukin-4 and Interleukin-5 in Nodular Prurigo and Pruritic Papular Lesions Ayu Wikan Sayekti, Ann Kautsaria Putri, Dwi Retno Adi Winarni, Satiti Retno Pudjiati Folia Medica Indonesiana.2024; 60(1): 47. CrossRef
Challenges and barriers to e-leadership participation: Examining the perspectives of Malaysian secondary school teachers Cha Shi Ping, Lokman Mohd Tahir, Mohd Shafie Rosli, Noor Azean Atan, Mohd Fadzli Ali Education and Information Technologies.2024; 29(8): 10329. CrossRef
One-Leg Standing Test with Eyes Open as a Screening Tool for Prefrailty in Community-Dwelling Older Japanese Women Zhenyue Liu, Shuji Sawada, Hisashi Naito, Shuichi Machida Healthcare.2024; 12(23): 2378. CrossRef
Recent Advancements in Subcellular Proteomics: Growing Impact of Organellar Protein Niches on the Understanding of Cell Biology Vanya Bhushan, Aleksandra Nita-Lazar Journal of Proteome Research.2024; 23(8): 2700. CrossRef
A comparison study: The use of digital and conventional impression techniques in dental hygiene education Raha K. Naderi, Tulsi J. Patel, Michelle A. Thompson Journal of Dental Education.2024; 88(5): 518. CrossRef
An Evaluation of the Nunez Community College Quality Enhancement Program: Assessing the Impact of Embedded Conover Workplace Readiness® Modules on Students’ Work Readiness Sandra Aurand Bosch, Donna Mowery Rice Sage Open.2024;[Epub] CrossRef
Integrating electric vehicles into hybrid microgrids: A stochastic approach to future-ready renewable energy solutions and management Aykut Fatih Güven Energy.2024; 303: 131968. CrossRef
Comparison of postural assessment and awareness in individuals receiving posture training using the digital AI posture assessment and correction system Musa Çankaya, Fatma Nur Takı International Journal of Occupational Safety and Ergonomics.2024; 30(4): 1311. CrossRef
Detection and Analysis of Fake News Users’ Communities in Social Media Abdelouahab Amira, Abdelouahid Derhab, Samir Hadjar, Mustapha Merazka, Md. Golam Rabiul Alam, Mohammad Mehedi Hassan IEEE Transactions on Computational Social Systems.2024; 11(4): 5050. CrossRef
How do fish miss? Attack strategies of threespine stickleback capturing non-evasive prey Seth Shirazi, Timothy E. Higham Journal of Experimental Biology.2024;[Epub] CrossRef
Effects of Occlusal Contact on Maxillary Alveolar Bone Morphology in Patients with and without Anterior Open Bite: A Cross-Sectional Study Chiyo Shimizu-Tomoda, Yuji Ishida, Aiko Ishizaki-Terauchi, Yukari Mizoguchi, Shuji Oishi, Takashi Ono Journal of Clinical Medicine.2024; 13(11): 3061. CrossRef
Bioceramic modular tissue-engineered bone with rapid vascularization for large bone defects Siwei Luo, Zhen Wang, Jialin He, Geng Tang, Daizhu Yuan, Zhanyu Wu, Zihao Zou, Long Yang, Tao Lu, Chuan Ye Ceramics International.2024; 50(11): 18275. CrossRef
Effect of Arogya Raksha Panchatantra (five lifestyle principles) on heart rate variability, menstrual symptoms, health-related quality of life, performance and self-efficacy in Young female adults with primary dysmenorrhea: protocol for an exploratory ran Karishma Silwal, Prakash Babu Kodali, Vakeel Khan, Hemanshu Sharma, Gulab Rai Tewani, Pradeep M. K. Nair CCRYN Indian Journal of Yoga & Naturopathy.2024; 1(1): 15. CrossRef
Letter to the editor on: “Calcaneal positioning in equinus immobilization of the ankle joint: A comparison of common orthoses used in the treatment of acute Achilles tendon ruptures” Rahul Kumar, Diggaj Shrestha, Kunal Setia, Sunita Sharma Foot and Ankle Surgery.2023; 29(6): 498. CrossRef
Irradiation damage reduces alloy corrosion rate via oxide space charge compensation effects Zefeng Yu, Elizabeth Kautz, Hongliang Zhang, Anton Schneider, Taeho Kim, Yongfeng Zhang, Sten Lambeets, Arun Devaraj, Adrien Couet Acta Materialia.2023; 253: 118956. CrossRef
Predictors of Relevant Changes in Pain and Function for Adolescents With Idiopathic Scoliosis Following Surgery Samia Alamrani, Adrian Gardner, Alison B. Rushton, Deborah Falla, Nicola R. Heneghan Spine.2023; 48(16): 1166. CrossRef
A multicenter feasibility study on implementing a brief mindful breathing exercise into regular university courses Annika C. Konrad, Veronika Engert, Reyk Albrecht, Christian Dobel, Nicola Döring, Jens Haueisen, Olga Klimecki, Mike Sandbothe, Philipp Kanske Scientific Reports.2023;[Epub] CrossRef
Adult children of parents with mental illness: Family stigma and coping on sense of self Chynna Campbell, Pamela Patrick Child & Family Social Work.2023; 28(3): 622. CrossRef
Veri Madenciliğine Dayalı Olarak Çalışanların Örgütsel Bağlılık Düzeyinin Belirlenmesi: İstanbul ve Kocaeli Örneği Nadir ERSEN, Timuçin BARDAK, Uğur Can USTA Bartın Orman Fakültesi Dergisi.2023; 25(3): 398. CrossRef
Pooled rates and demographics of POTS following SARS-CoV-2 infection versus COVID-19 vaccination: Systematic review and meta-analysis Shin Jie Yong, Alice Halim, Shiliang Liu, Michael Halim, Ahmad A. Alshehri, Mohammed A. Alshahrani, Mohammed M. Alshahrani, Amal H. Alfaraj, Lamees M. Alburaiky, Faryal Khamis, Muzaheed, Bashayer M. AlShehail, Mubarak Alfaresi, Reyouf Al Azmi, Hawra Alba Autonomic Neuroscience.2023; 250: 103132. CrossRef
Rapid suture-free repair of arterial bleeding: A novel approach with ultra-thin bioadhesive hydrogel membrane Siwei Luo, Long Yang, Qiang Zou, Daizhu Yuan, Shunen Xu, Yanchi Zhao, Xin Wu, Zhen Wang, Chuan Ye Chemical Engineering Journal.2023; 472: 144865. CrossRef
The Relationship Between Obsessive-Compulsive Disorder and Gaming Disorder Nazir Hawi, Maya Samaha International Journal of Cyber Behavior, Psychology and Learning.2023; 13(1): 1. CrossRef
Microbiological Quality and Presence of Foodborne Pathogens in Raw and Extruded Canine Diets and Canine Fecal Samples Doina Solís, Magaly Toro, Paola Navarrete, Patricio Faúndez, Angélica Reyes-Jara Frontiers in Veterinary Science.2022;[Epub] CrossRef
Stapler Assisted Total Laryngectomy: A Prospective Randomized Clinical Study Omar Ahmed, Hesham Mustafa Abdel-Fattah, Hisham E. M. Elbadan Indian Journal of Otolaryngology and Head & Neck Surgery.2022; 74(S2): 2205. CrossRef
Project VLOGI (Video Lectures on Giving Instructions): Effects on Learners’ Performance in Probability and Statistics Sherwin BATİLANTES International Journal of Educational Studies in Mathematics.2021; 8(4): 299. CrossRef