Statistical Measurements

The purpose of this assignment is to write a paper including a comparison table. The table
will be provided as an appendix to the final paper.
locate a quantitative, qualitative, and mixed methods article related to your DPI topic.
(Educating type 2 diabetes patients on the importance of medication adherence).
You may reuse articles from order #115442 paper
Using the “Comparative Table Template,” create a table that compares quantitative,
qualitative, and mixed methods articles. Provide the table as an appendix to the paper.
Organize the table according to the following.

  1. Type of article.
  2. Title of the article.
  3. Type of analysis: Describe the process of analysis. There can be more than one.
  4. Applicability of test: Describe other analyses that could be used and why?
  5. Reliability and validity: Describe either the reliability and validity measures of tools used
    in the study or the techniques used to enhance reliability and validity in the study.
    Write a 1,000 word paper discussing the application of data analysis, reliability and
    validity, and assorted statistical tests used in health related research. Include the following
    in your paper:
    State the types of study used in each article.
    Discuss the types of statistical tests used within each article and why they have been chosen.
    Discuss the applicability of the chosen statistical test and why the statistical test was chosen.
    Discuss the differences between parametric and nonparametric tests and how they were
    applied in the articles you chose.
    Evaluate how the factors of reliability and validity are accounted for in the articles (of
    instruments and surveys or in the design and data collection conducted).
    Summarize how the chosen studies could be applied within the context of your practice.


Statistical Measurements

Types of Study in Each Article
The first study is a case-control quantitative type of study. Type two diabetic patients on
different therapies were included in the study through a diabetic clinic. The patients were
assessed using semi-structured and face-to-face interviews to obtain their views on the
importance of nutritional therapy about therapies (Pinto, Braz, Nascimento and Eurico Gomes,
The second study used a case-control qualitative study. Diabetic patients were assessed
on their illness experiences and disease management. The assessment was done using interviews
(Graffigna, Barello, Libreri, & Bosio, 2014).
A longitudinal cohort study with mixed methods was applied in the third study. The
effectiveness of using online oriented self-management by type two diabetes patients was
assessed. Patients were followed up throughout two weeks to evaluate the impact (Hofmann,
Dack, Barker, & Murray, 2016).
Types of Statistical Tests Used
Mann-Whitney test was applied in the first study to establish the relationship between the
quality of diet and patient demographics. The test tried to find out how sex, age, and duration of
illness affected the quality of diet the patients adopted. Mann –Whitney test is vital because it
helps identify how different variables affect each other (Corbert, Sibbald, Stockton, & Wilson,


ANOVA test was also used to analyze the importance of physical activity concerning
dietary and pharmacological therapies. The test was also used to establish the relationship
between the body mass index and food choices. The t-test was used to evaluate the differences in
treatment modalities (Corbert et al., 2015).
Study two relied on interpretive analysis to analyze the data. Associations between
different variables were identified and a more in-depth explanation of findings established. This
provided an avenue of understanding different patient actions. This type of statistical test helps to
understand the behavior of patients. Explanations of how people operate, understand situations
and the kind of actions they take is realized through this test (Smith, Williams, & Oswald, 2013).
In the third study, Shapiro-Wilk test was applied which is vital to identify relations from
normal parameters. Mann –Whitney test was also used to test quantitative data where pre and
post outcome measures were analyzed. Qualitative data were analyzed using interpretive content
analysis method where different patient experiences were identified (Algina, 2017).
Applicability of the statistical
In the first study, ANOVA analysis indicated a lack of significant correlation between
dietary intake and the patients’ weight: normal, excess and obese. There was also no statistically
significant correlation between perceptions of the quality of diet about demographic
characteristics of patients according to the Mann – Whitney test. Bonferroni’s correction and t-
test combined indicated a lack of significant variations in the importance attached to each type of
therapy (Smith, Williams, & Oswald, 2013).


The attitudes of patients in the second study were analyzed under different spheres of
their daily life for example diet, exercise, and therapy and nurse-patient relationship. The
responses on these particular areas were coded and a measure of central tendency calculated. The
responses were also used to come up with conclusions of the study (Johnson, Kording, Hargrove,
& Sensinger, 2017).
In the third study, Shapiro Wills test (0.05) implied that there was a deviation of PAID
and HADS from normal parameters. Pre and post outcome measures were analyzed using the
Mann Whitney test. There was a lack of noticeable variations between the two; an indication that
the Internet-based self-management had an impact on the patient’s psychosocial wellbeing
(Johnson et al., 2017).
Differences between Parametric and Nonparametric Tests
Parametric tests include t-test, f-test, z-test, and ANOVA test while non-parametric tests
include the Mann Whitney test, Rank sum test and Kruskal Wallis test. A combination of
different tests was used in each study to analyze the study outcomes (Slack, 2014).
Parametric tests are applied in testing studies where the data about the study population is
well known while in the case of nonparametric test, use the study population is not well known.
In these studies, parametric tests applied included ANOVA test, t-test, and Shapiro Wills test. In
the first study ANOVA tested the importance attached to different therapies by patients and t-test
helped identify the differences in the therapies. Mann Whitney was also used in this test to
establish the relationship between quality of patient diet and demographic characteristics
(Ihantola, & Kihn, 2015).


The parametric test is applied in a study when assumptions on the distribution of data can
be made. This is because the subject population is well known and there is much information on
the variables where assumptions are being made. When the population information is limited, no
assumptions about data distribution can be made, and therefore nonparametric tests are applied.
For instance, in the third study, Shapiro Wilk test was used to identify deviations from normal
parameters. Mann Whitney test was applied in the analysis of pre and post outcome measures
(Smith, Williams, & Oswald, 2013).
Parametric tests use the mean of the results to analyze data while nonparametric tests use
the median. Therefore, a nonparametric test is applied when the original hypothesis does not
match the data. The Shapiro Wilk test applied in the third study, the mean of the results were
calculated and deviations identified while the Mann Whitney test applied used the median to
establish the deviations (Corbert et al., 2015).
Accountability of Reliability and Validity
The first study was able to measure what it was aimed at measuring. However, the results
are not conclusive as the information obtained from patients was based on recall memory.
Patients may not be able to remember what they ate two weeks ago. Other factors may affect
weight gain that is not diet related, for example, the presence of co-morbid illnesses and genetic
factors. The use of daily diet intake diaries over a period will help eliminate the recall bias in the
study (Clemens & Demombynes, 2013).
It is impossible to apply the findings of the study to a broader population because the
study sample size used was small. Using a bigger sample size can give a real picture of the


broader population. The study did not examine the participant’s views, and how they affect the
choice of foods, they take (Nayak, 2013).
Combining interviews with questionnaires and patient diaries in the second study helped
reduce the error associated with the recall memory that majorly affected the first study during
data collection. Patients may not be able to recall the details for the meals taken two weeks ago
and may give the wrong information to satisfy the interviewer. However, the snowball technique
used in sampling may also bring the aspect of bias as the researcher may end up obtaining
similar information from participants, which may not reflect the real situation (Johnson et al.,
The qualitative design applied in the second study helped capture all the aspects of
psychosocial phenomena. The narrative form adopted encouraged patients’ self-reflection
leading to the realization of more patient experiences about their condition. The use of
interpretive content analysis method provided a wide range of factors that hinder and encourage
patient engagement despite the use of a small sample size (Nayak, 2013).
Use of mixed methodology in the third study helped to capture all aspects of the research
problem. The qualitative aspect through interviews helped in capturing personal opinions about
the condition as well as the effectiveness of the program. The quantitative approach helped in
comparing more demographics of participants in a detailed manner. The two methods
complemented each other ensuring the success in achieving validity (Ansari, Harris, Zwar, &
Hosseinzadeh, 2017).


The sample size calculation in the third study was done with the help of G Power using
an alpha of 0.05. This lead to the achievement of the approximately minimum size required
being useful clinically (Ansari et al., 2017).
In conclusion, the findings from the three studies can be used to develop strategies, which
ensure that type two diabetic patients receive all aspects of care. The nurses should educate
patients on the importance of all treatment modalities and the effect of their synergy on their
health. Minimizing disengagement will increase adherence to treatment hence the success in
controlling diabetes type two among patients. The care should take into consideration the
psychosocial aspect of care.



Algina, J. (2017). Type I error rates and power estimates of selected parametric and
nonparametric tests of scale. Journal of Educational Statistics, 12(1), 45-61.
Ansari, R. M., Harris, M. F., Zwar, N. A., & Hosseinzadeh, H. (2017). A Quantitative Research
on Self-management of Type 2 Diabetes.
Clemens, M. A., & Demombynes, G. (2013). When does rigorous impact evaluation make a
difference? The case of the Millennium Villages. Journal of Development Effectiveness,
3(3), 305-339.
Corbett, N., Sibbald, R., Stockton, P., & Wilson, A. (2015). Gross Error Detection: Maximising
Graffigna, G., Barello, S., Libreri, C., & Bosio, C. A. (2014). How to engage type-2 diabetic
patients in their health management: implications for clinical practice. BMC public
health, 14(1), 648.
Hofmann, M., Dack, C., Barker, C., & Murray, E. (2016). The impact of an Internet-based self-
management intervention (HeLP-diabetes) on the psychological well being of adults with
type 2 diabetes: a mixed-method cohort study. Journal of diabetes research, 2016.
Ihantola, E. M., & Kihn, L. A. (2015). Threats to validity and reliability in mixed methods
accounting research. Qualitative Research in Accounting & Management, 8(1), 39-58.
Johnson, R. E., Kording, K. P., Hargrove, L. J., & Sensinger, J. W. (2017). Adaptation to random
and systematic errors: Comparison of amputee and non-amputee control interfaces with
varying levels of process noise. PloS one, 12(3), e0170473.
Nayak B. (2013). Understanding the relevance of sample size calculation: Indian J.
Ophthalmolol, 58, 469-470.


Pinto, Braz, Nascimento1 and Eurico Gomes, (2017). Do Patients Value Nutritional Therapy? A
Quantitative Study in Type- 2 Diabetes Patients: Int J Diabetes Clin Res, 4 (079), 2377-
Slack, M. K. (2014). Establishing the internal and external validity of experimental studies.
American Journal of Health-System Pharmacy, 58(22), 2173-2181.
Smith, B. L., Williams, B. M., & Oswald, R. K. (2013). Comparison of parametric and
nonparametric models for traffic flow forecasting. Transportation Research Part C:
Emerging Technologies, 10(4), and 303-321.
The Use of Data with UBA on Global Producer III (Part 2). 33rd International North Sea


Type of

Title of Analysis and

Applicability Reliability and


Do Patients Value
Therapy? A
Quantitative Study
in Type-
2 Diabetes Patients

Mann Whitney
test was used to
establish the
between two
ANOVA and t-
test was used to
compare the

According to
Mann Whitney
test, there is no
dietary intake and
ANOVA test
showed that
patient weight is
not affected by
the dietary intake

The use of recall
memory during
data collection
may lead to
obtaining of false
The sample size
saw relatively
little hindering
transferability of
the findings.


How to engage
type-2 diabetic
patients in their
own health
implications for

An interpretive
content analysis
was applied to
obtain a wide
range of patient

Different spheres
of patient life
were identified
and their
attitudes towards

Involved use of
relatively small
sample thus
cannot be

Methods The Impact of an
Internet-Based Self-
(HeLP-Diabetes) on
the Psychological
Well-Being of
Adults with
Type 2 Diabetes: A
Cohort Study

Mann Whitney
test was used to
establish a
between pre and
post outcome

deviated from
normal as per
Shapiro Wilk test.

mixed methods
helped to capture
all aspects of the
topic of study.
Results cannot be
because of the
limitations that
were associated
with the