Research Methods: Statistical Analysis and Interpretation

This assignment addresses several key aspects of research methodology, including statistical analysis and interpretation. The solution begins by explaining the application of z-tests, interpreting the results, and discussing the limitations in non-normal distributions. It then delves into regression analysis, interpreting p-values to determine relationships between variables like blood pressure, weight, age, and sex. The assignment also covers the interpretation of R-squared values and constant terms. Further, the solution elucidates the differences between effect estimation and hypothesis testing, the importance of precision and statistical significance versus clinical significance, and the role of critical values, sample size, and power in research. The provided references include relevant books and journals supporting the analysis. The assignment covers topics such as bias, randomized clinical trials (RCTs), data distributions, and correlation analysis, providing a comprehensive overview of research methods and statistical concepts.

Research Method questions

Table of Contents
5..............................................................................................................................................2
6..............................................................................................................................................2
7..............................................................................................................................................2
REFERENCES...........................................................................................................................4

5
a) z-test is carried out and it is valid statistical test because the number of n is greater than 30.
b) As per the above the z test value is -14.29 which is lower than 0.05 thus alternative
hypothesis is used that shows that there is a direct relationship between blood pressure and
other variables.
c) This is not applicable in the non-normal distribution because it is appeals to the Central
limit theorem. Also, the method take enough time for the coding and as a result, most of the
researcher do not use this test.
6
a) By applying the regression analysis, it is interpreted that most of blood pressure is depend
upon some factors like, as per the weight the p value is 0.276 which is higher than 0.05 and as
a result, null hypothesis is accepted. Therefore, there is no relationship between BP and
weight, while there is a strong relationship between age, cons with blood pressure because the
p value is 0.00 which is lower than 0.05 and as a result, alternative hypothesis is accepted.
Also, the relationship between sex and blood pressure is low because the p value signifies
0.007 which is greater than 0.05 and that is why alternative hypothesis is accepted.
b) Constant term in the regression is the evidence which an individual with non- zero values
of all the predictor is expected to have non zero response. As per the table, the constant is
0.00 which shows that there is a direct relationship with the blood pressure as the value of
greater than 0.05.
c) The R- squared value represented that the first value is associated with other but it is
influence with the same value as well. Like, as per the table, the value is 0.40, when there is
an additional explanatory covariates were added then the value is influenced by 0.40 from the
actual results.
d) From the table it is analyzed that constant term has more than 95% of the confidence then
this value is confident that the selected sample mean for the specified values of the variable in
the model while other variables are not.
7
Effect estimation versus hypothesis testing: Through estimation testing researcher
determine the 95% level of confidence interval instead of using P values while, hypothesis

testing is set up about the parameter by rejecting and accepting the statement (Washington
and et.al., 2020).
Precision of estimates: This help to determine the close estimation of each sample
from each other i.e. using standard deviation to determine the precision instead of using p
value.
Statistical significance versus clinical significance: Statistical significance helps to
determine the reliability of study results, while clinical significance used to interpret the
results which are being clinically important by using specific key terms.
Critical values (Level of significance): It is demoted as α which is a critical mean
value such that If this value is less than 0.05 then alternative hypothesis is accepted and if not
null hypothesis is accepted.
Sample size: It means that if majority of the people present their views for a
particular topic, then the results are in favor of it (Du and et.al., 2020).
Power: This help to directly related to test of hypothesis and it is used when there is a
small size of sample that help to detect the effect of given test at desire level of significance.

REFERENCES
Books and Journals
Du, A. and et.al., 2020. Statistical estimates of hominin origination and extinction dates: A
case study examining the Australopithecus anamensis–afarensis lineage. Journal of
human evolution, 138, p.102688.
Washington, S. and et.al., 2020. Statistical and econometric methods for transportation data
analysis. CRC press.