What demographic variables were measured at the nominal level of measurement in the Oh et al. (2014) study?
Nominal measurements are the type of data which is not measured by numbers but presented in the form of different variants (for example, “yes” or “no” options, “male” or “female” gender, or a choice among various possibilities). In the Oh et al. (2014) study, the following nominal demographic parameters were measured: fractures occurrences; drinker or non-drinker, smoker or non-smoker, regular exercises practices presence. According to the mineral density of bones, there were “normal,” “osteopenia,” and “osteoporosis” categories.
What statistics were calculated to describe body mass index (BMI) in this study? Were these appropriate?
To determine the Body Mass Index (BMI), the information about person’s height and weight is needed. Traditionally, height is measured in meters and weight – in kilograms. For the BMI calculation, weight should be divided by height. In this investigation, height was measured in centimeters. However, this measurement is easily transformed in meters by dividing by 100. The weight was measured in kilograms. In the study, calculation of the BMI was performed correctly. It was estimated that there was not the significant difference between intervention and control BMI values
Were the distributions of scores for BMI similar for the intervention and control groups?
It was estimated that the score of BMI was similar for intervention and control groups. Mean BMI for the intervention group was 24.7 (SD = 2.5); mean BMI for the control group was 23.38 (SD = 3.32). Therefore, it could be stated that two means were overlapped. Also, the t score and the chi-squared score were not significantly different. The calculated value of p was 0.485 while it was required that p should be 0.05 or less for obtained results to be significant.
Was there a significant difference in BMI between the intervention and control groups?
According to the study results, there was no statistically significant difference between BMI scores for the intervention and control groups regarding the baseline characteristics. It could be proved by the value of p. It was calculated that in the study p=0.485 for the calculation with 40 degrees of freedom. This value is considered as non-significant.
Based on the sample size of N = 41, what frequency and percentage of the sample smoked? What frequency and percentage of the sample were non-drinkers (alcohol)?
Generally, 41 persons were examined (the sample size is 41). In Table 2, the information about participants was presented. Among all 41 participants, 40 were non-drinkers. Therefore, the percentage of non-drinkers was the following: 40/41*100=97.6%. Among all the examined participants, all 41 were non-smokers. Therefore, the percentage of non-smokers was the following: 41/41*100=100%.
What measurement method was used to measure the bone mineral density (BMD) for the study participants?
To measure the bone mineral density (BMD), a DEXA scanning method was applied. This approach is widely used in medical investigations. It was claimed that it has high-quality standards. In the Oh et al. (2014) study, it was postulated that the rate of error of measurements was less than 1%.
What statistic was calculated to determine differences between the intervention and control groups for the lumbar and femur neck BMDs? Were the groups significantly different for BMDs?
The mean of the BMD score was calculated. It was used to determine that the difference between BMD scores of intervention and control groups existed. To estimate the statistical difference between the scores, the t-statistic criterion of the difference of means was calculated. A T-statistic criterion was estimated with degrees of freedom. It was determined that the t-value was 0.0526, and the p-value for this result was 0.958. For t=0.055, the value of p is 0.956. These values of p are considered as non-significant because the maximum significant score of p is 0.05.
The researchers stated that there were no significant differences in the baseline characteristics of the intervention and control groups (see Table 2). Are these groups heterogeneous or homogeneous at the beginning of the study? Why is this important in testing the effectiveness of the therapeutic lifestyle modification (TLM) program?
Participants for both groups (intervention and control) were selected randomly. It was important because groups for this study are needed to be homogeneous. Baseline characteristics both groups were determined. It was stated that these parameters were not significantly different for intervention and control group. This procedure was essential for testing TLM. If two groups were not homogeneous, the differences found in BMD scores could be connected to the groups’ difference but not to the TLM intervention.
Oh et al. (2014, p. 296) stated that “the adherence rate to the TLM program was 99.6%.”
According to the SPIRIT statement, the statistical significance of obtained results and, therefore, their interpretation could be influenced by the adherence. Low or no adherence leads to the low statistical power of investigation’s results. Besides, the reduction between studies could appear. Results of such investigation could be considered as unreliable. In the Oh et al. (2014) study the adherence was 99.6% which make results of the investigation reliable.
Was the sample for this study adequately described?
In the study, the following sample characteristics were provided: weight, height, the BMI, age, smokers or non-smokers, drinkers or non-drinkers, fraction history. It could be stated that the sample description was not complete. Other essential characteristics could be provided. For example, it is important to know the race of participants. It is also a baseline feature which might affect the BMD. The relevant information according to the groups’ sample description was also missed. There were no data about procedures that were performed by the control group instead of the studied intervention. These activities might also affect the investigation’s results.
Oh, E. G., Yoo, J. Y., Lee, J. E., Hyun, S. S., Ko, I. S., & Chu, S. H. (2014). . Research in Nursing & Health, 37(4), 292–301. Web.