Question 1. Does any difference exist in BMD estimates at the femoral neck between people aged over 50 compared to those under 50
Research Question
Are there differences between the femoral-neck BMD in persons >50 compared to the femoral-neck BMD in 0 to 50 persons?
Research Hypothesis
“Null Hypothesis (H0)”: There is no difference in the averageness of “femur-neck BMD of people aged 50” or older and people aged over 50 or older larger between age 50 and older and age more than 50 years or older.
“Alternative Hypothesis (H1)”: There is no equal mean of the BMD of the femoral-neck between subjects age 50 years old and less, in one age group (under age 50 years old), and in another age group (age of older age than 50 years old).
Analysis
“Bivariate and univariate analysis” are conducted to characterize variables and to estimate relationship between “age groups” and the “femoral-neck BMD (DXXNKVMD)” (O'Rourke et al. 2021).
Understanding statistical methods, data interpretation, and hypothesis testing is important for students working on research-based assessments. Students seeking academic guidance can explore best assignment help Australia to better understand statistical concepts and assignment structure.
Univariate analysis
“Univariate analysis” generalizes and describes the general properties of variables included in the “data set, including, Age 2grps and "Femoral neck BMD"”.
Description of each variable
Age2grps: is a grouping variable that divides the participants into two groups; older or younger than 50 years of age (>50, 50 years, or older, 50 years). “"Femoral neck BMD (DXXNKVMD)” is also a “continuous dependent variable” that represents the mineral density of “femur-neck bone (BMD)”.
Details of the above “categorical variables”
The “Age variable” is categorical Age 2grps because the sample can easily be categorized and the comparison between -BMD of younger and older individuals would be easier (Wáng & Xiao, 2022).
Graphical representation of each categorical variable
(Source: Self-created in Jamovi)
A “bar diagram” was designed to represent the balance of the participants in the two age groups (Li, et al. 2023).
Characterization of the continuous variables in more detail
“Femoral neck BMD (femoral-neck bone mineral density)” is such kind of dependent variable that provides readings of the bones mineral density of each subject (Nguyen et al. 2025).
The number of datasets and the missing values, with descriptive statistics (minimum, maximum, SD, median, mean)
(Source: Self-created in Jamovi)
“Continuous variable "Femoral neck BMD"”: N = 3000: no levels given as missing, but extreme sentinel values distort the results (mode = 9999, max = 9999, range = 9999). These attributed the “mean (2077) and the SD (4056)” but the median (0.881)” is the only coefficient that displays true central tendency.
The values of IQR are described.
The “Quartile Range (IQR)” of the continuous variable, the "Femoral neck BMD" = 0.339; that is the dispersion joining the midpoint parting of the information. The value orientating towards the reduced IQR would denote concentration of the majority of the BMD more or less around the median and an enhanced value ought to denote a widened spread (Xue et al. 2021).
Normal distribution of the continuous variables
(Source: Self-created in Jamovi)
Table of normality check
(Source: Self-created in Jamovi)
Graphical representation
(Source: Self-created in Jamovi)
Bivariate Analysis
(Source: Self-created in Jamovi)
Assumptions and statistical test
(Source: Self-created in Jamovi)
The :Shapiro-Wilk ( p <.001)” suggests that "Femoral neck BMD" is highly “non-normative” and therefore fails in “t-test normality”.
(Source: Self-created in Jamovi)
Here, the “Levene test (p = 0.005)” is interesting, in fact, the equality of the variance is not observed.
(Source: Self-created in Jamovi)
The differences in the “means (0.930 vs 0.815)” and extraordinarily high SDs are also indicators of offset measures of central tendency most likely attributable to “outliers (or sentinel values)” in the data.
(Source: Self-created in Jamovi)
(Source: Self-created in Jamovi)
Clinical implications of the results
Effect size
The “standard effect size of Cohen d” proved to be -0.0509 and reflects an insignificant variance in the “female-neck BMD” between age groups (Li et al. 2025).
Direction of effect
The “Mean of BMD” in the category of over 50 (2180) was marginally higher than under 50(1974). Interestingly the median BMD of the groups did not repeat the same trend, however, the median BMD of the and far older group, in the Under-50 range less than that of the Over-50 group was less by 2.8, showing that; extreme outliers had/could have distorted the mean.
Other forms of confounding variables.
The “gender, physical activity” and intake of “calcium and vitamin D” in the diete, creation of drugs and the genetic susceptibility is important in the BMD.
Detailed of the statistical results of both “bivariate and univariate analysis”
“Shapiro-Wilk (p < 0.001)” confirmed that the “continuous variable "Femoral neck BMD"” was not distributed normally, and “Levene (p = 0.005)” indicated that the groups did not have the same variance levels of the variable. Therefore, “Mann-Whitney U test” was utilized that revealed that BMD had a significant difference “(p < 0.001)” among the age groups.
Results clinical implications entitled to analysis of issues
Whereas the results suggest an actionable change in the BMD between patients older than 50 and younger than 50 years, the effect size is very small, which implies a low clinical impact. These findings support the argument of the “multifactorial screening” rather than the intention of basing on the view of age to one determinant of bone health.
Understanding descriptive statistics, hypothesis testing, and data interpretation is essential when preparing a Statistics Assignment involving research findings and evidence-based conclusions.
Reference List
Journals
- Li, H., & Wang, H. (2025). Association between weekend warrior physical activity pattern and bone mineral density among adults: national health and nutrition examination survey. Osteoporosis International, 1-9. [Retrieve from: https://link.springer.com/article/10.1007/s00198-025-07535-9] [Retrieve on: 01.09.2025]
- Li, S., Li, L., Feng, A., Huang, T., Chen, C., He, N., ... & Lyu, J. (2023). The role of hypertension in bone mineral density among males older than 50 years and postmenopausal females: evidence from the US National Health and Nutrition Examination Survey, 2005–2010. Frontiers in Public Health, 11, 1142155. [Retrieve from: https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2023.1142155/full] [Retrieve on: 01.09.2025]
- Nguyen, H. G., Nguyen, D. T., Tran, T. S., Ling, S. H., Ho-Pham, L. T., & Van Nguyen, T. (2025). Artificial intelligence system for predicting areal bone mineral density from plain X-rays. Osteoporosis International, 1-10. [Retrieve from: https://link.springer.com/article/10.1007/s00198-025-07634-7] [Retrieve on: 01.09.2025]
- O'Rourke, D., Beck, B. R., Harding, A. T., Watson, S. L., Pivonka, P., & Martelli, S. (2021). Assessment of femoral neck strength and bone mineral density changes following exercise using 3D-DXA images. Journal of Biomechanics, 119, 110315. [Retrieve from: https://www.sciencedirect.com/science/article/pii/S0021929021000956] [Retrieve on: 01.09.2025]
- Wáng, Y. X. J., & Xiao, B. H. (2022). Estimations of bone mineral density defined osteoporosis prevalence and cutpoint T-score for defining osteoporosis among older Chinese population: a framework based on relative fragility fracture risks. Quantitative Imaging in Medicine and Surgery, 12(9), 4346. [Retrieve from: https://pmc.ncbi.nlm.nih.gov/articles/PMC9403581/] [Retrieve on: 01.09.2025]
- Xue, S., Zhang, Y., Qiao, W., Zhao, Q., Guo, D., Li, B., ... & Yang, S. (2021). An updated reference for calculating bone mineral density T-scores. The Journal of Clinical Endocrinology & Metabolism, 106(7), e2613-e2621. [Retrieve from: https://academic.oup.com/jcem/article-abstract/106/7/e2613/6178347] [Retrieve on: 01.09.2025]
