Qualitative and Quantitative Studies

Part 1
Continuous Versus Categorical Variables
A continuous variable is any variable that can take any value between two numbers
(Shieh, 2019). For example, in a study, height is a continuous variable. When dealing with
twenty participants between 70 and 90 inches, there are many values in between. It is a numeric
variable with infinite values or limitless possibilities between any two figures or values. A
participant can be 60.8976 inches tall, while another maybe 80.3456 inches. A continuous
variable is one that is measurable. These variables also have a logical order. For instance, it is
possible to arrange participants in order of height from the shortest to the tallest. Other examples
include temperature and weight.
On the other hand, a categorical variable has a finite number of distinct groups. Data used
with these variables lack a logical order (Shieh, 2019). A categorical variable involves non-
numerical categories. Such a variable can only be presented in a fixed number of values. For
example, some of the categorical variables used include gender. One may either be male or
female. When a numerical value is used, it does not have any numerical significance. For
instance, in a study, a researcher may use 1 to represent females and 2 to represent males. Here,
the values used are indicators for easy analysis when amazing data and 2 does not mean that the
male gender is stronger or more in numbers compared to the female, which is identified using
numeral 1.
Nominal Data versus Interval Data

Nominal data refers to data that can only be classified or labeled into categories.
Categories in nominal data cannot be ranked or labeled in any meaningful way (Nazabal et al.,
2020). For example, when studying means of transportation as a nominal variable, the nominal
data here would include different classes or modes of transport. One category may have buses,
another trains, bicycles, while the last one may have cars. Whichever way one decides to present
them, the order is meaningless. For instance, if trains appear first on the list, it does not mean that
they are better means of transport compared to bicycles and vice versa.
Interval data, on the other hand, can be measured, ranked, categorized, and there exist
even spaces between any two values (Nazabal et al., 2020). Thus, the distance between two
adjacent values is equal. The distance is called intervals. The interval scale does not have a true
zero value. Thus, zero in interval data does not imply lack or nothingness. Zero does not mean
complete absence but is rather an arbitrary point (Bandalos, 2018). Temperature as a variable fall
under this category of data. Zero degrees does not mean an absence of temperature. Additionally,
zero degrees is not the lowest possible level of temperature. Further, other than lacking a true
zero value, values in interval data cannot be divided or multiplied. For example, 20°C is not
twice as hot as 10°C.
Part 2
Qualitative Article
Lawrence et al. (2019) conducted meta-analyses to evaluate the risk associated with anxiety
disorders among children born and raised by parents who have a history of anxiety disorders.
The researchers were interested in establishing whether there existed evidence showing that
when parents have a history of a specific anxiety disorder, there is a specificity of risk in that the

offspring will also suffer from the particular anxiety disorder. The researchers examined whether
the risk of anxiety disorders for offspring was moderated by a child’s gender, age, temperament,
and the existence of depressive disorders in a child’s parents.
The researchers searched for similar studies conducted on their topic of interest. They
searched in various databases such as Web of Science, PubMed, and PsycINFO. The search was
conducted in June 2016 and July 2017 (Lawrence et al., 2019). It was a one-year study. The
researchers also had a set criterion. Any article that was selected for the study had to meet the
criteria to be eligible for analysis. They were interested specifically in articles that were
published in peer-reviewed journals and thus wanted articles that were highly credible. Further,
the article had to contain at least one group of parents that had anxiety disorders, which was
being compared to another group that did not have a history of anxiety disorders. The researchers
also wanted articles that had used diagnostic tools that had already been validated in their study.
The aim was to ensure that the results were reliable. The offspring included in the study must
have had reported cases of anxiety disorder prior to the study. In total, 53 articles met the criteria
and were analyzed during the study (Lawrence et al., 2019).
The researchers used random and mixed-effect models in the study. They used the random-
effects meta-analyses to determine the risk ratios for diagnostic outcomes. Specifically,
Lawrence et al. (2019) used Knapp and Hartung adjustment due to the fact that existing studies
show the use of random-effects modeling meta-analysis in studies where there is a small number
of studies or where the variance heterogenous tends to increase the possibility of committing a
type 1 error. Further, the researchers determined the impact of heterogeneity of the effect of sizes
existing between the different studies used in the study using the I 2  statistic (Lawrence et al.,
2019). Continuous variables in the study include the proportion of male to female offspring

respondents, age of offspring, and rate of depressive or anxiety disorders in parents. The
researchers used the mixed-effects meta-analyses to evaluate the effects (moderation effects) of
these continuous variables (Lawrence et al., 2019).
Further, the researchers compared their results with results from other meta-analyses
conducted previously. For instance, they compared their results to a similar study conducted by
Micco et al. (2009) and established that the findings were consistent. The two studies concluded
that children whose parents have a history of anxiety disorders are also at a higher risk of
experiencing both depressive and anxiety disorders. The study also added or extended the study
by Micco et al. (2009). Lawrence et al. (2019) established that the risk is higher for depressive
disorders compared to that of anxiety disorders. Thus, there is a higher likelihood that a child
will experience depressive disorders if the parent also experienced the same. The risk for such a
child is higher than that of a child whose parent had anxiety disorders.
Overall, the study focused on participants who confessed diagnoses of depression and
anxiety. Therefore, these were the predictor, moderator as well as outcome variables assessed in
the study (Lawrence et al., 2019). The researchers were only able to conduct moderation
analyses for the sex and age of the offspring that took part in the study. They also analyzed
parents’ history with depressive disorder. The meta-analyses concluded by supporting the
provision of targeted prevention interventions and measures. If a parent has a history of anxiety
and depressive disorders, then measures should be taken to prevent the same from manifesting in
their children, who face an increased risk.
Quantitative Study

Lin et al. (2022) conducted a cohort study where 5507 participants were involved in the
study. The researchers were interested in understanding the detrimental impact associated with
depression and obesity and how it eventually transitions to functional disability. They conducted
a study where they investigated the connection between baseline depression-obesity status and
how it then eventually changes with the occurrence of functional disability. Their interest was in
Chinese people, specifically middle-aged and older individuals. Specifically, the participants
were 45 years or older. The study was longitudinal, and it took place from 2011 to 2015. The
5507 participants were cross-categorized, with a status of depression-obesity being the baseline.
The variables assessed during the study include obesity, depression, gender, age,
education level, systolic blood pressure, height, weight, and marital status at the time of the
study. The two main variables were depression and obesity. Depression was assessed using the
10-item Centre for Epidemiological Studies Depression Scale (CESD-10). The scores on the
CESD-10 scale ranged from 0-to 30, with a cut-off score of 10. A score of 30 on the scale
indicated severe depressive symptoms. Depression study is a continuous variable. The
researchers considered scores between 10 and 30. The scores on the scale indicated the severity
of symptoms, with 10 indicating the least severe symptoms and 30 indicating the most severe
depressive symptoms. However, the baseline status for both depression and obesity was
categorized as neither depression nor obesity, meaning that a participant did not have any of the
two conditions. Another category was depression without obesity, where a participant was only
depressed and not obese. There was also obesity without depression. The last category under the
baseline status was comorbidity where a participant had both depression and obesity. Based on
this status, this was a categorical variable. A participant could only belong to one category based
on their status.

Further, height and weight measurements were also taken. These helped in calculating the
BMI (body mass index) of the participants. The BMI was calculated as weight divided by height.
Any individual whose BMI fell at 28kg/m2 or higher was considered obese. Height and weight
are also continuous variables. Another variable that was assessed during the study is age. The
age of the participants was considered. Each had to be 45 years or older. Age is a categorical
variable. The researchers used age as a category where they classified the participants into
groups. Another categorical variable in the study is marital status. The researchers categorized
the participants as either married, not married, or cohabitated at the time of the study. Education
level was also assessed. This is a categorical variable, and so is gender, which is another variable
considered during the study. Participants were either male or female.
During the study, participants were categorized as either illiterate or those that never
received any formal education, those with primary school education, and another group was
those with middle school level of education or above. Another demographic factor that was
assessed is smoking status. Smoking status is a categorical variable. The researchers labeled
participants as either ‘ever smoker or never smoker’ (Lin et al., 2022). There was also the area of
residence where participants were categorized as either urban dwellers or rural dwellers. This,
too, is a categorical variable. The last categorical variable studied is drinking status. Respondents
were classified either as ‘ever or never drinker’ (Lin et al., 2022).
Further, Lin et al. (2022) measured systolic blood pressure and diastolic blood pressure.
The measurements were taken three times at equal intervals. This is a continuous variable. The
measurements taken were used to confirm the patient’s hypertension status. Another biochemical
measure that was taken during the study is multimorbidity, where participants were considered
multimorbid if they had two or more of these conditions; kidney disease, heart disease, asthma,

cancer, stroke, asthma, Alzheimer’s disease, and arthritis (Lin et al., 2022). Multimorbidity is a
categorical variable. Participants were in only one category, with a ‘yes or no’ label. Thus, one
could only belong to one category

References

Bandalos, D. L. (2018). Measurement theory and applications for the social sciences. Guilford
Publications.
Lawrence, P. J., Murayama, K., & Creswell, C. (2019). Systematic review and meta-analysis:
anxiety and depressive disorders in offspring of parents with anxiety disorders. Journal of
the American Academy of Child & Adolescent Psychiatry, 58(1), 46-60.
Lin, L., Bai, S., Qin, K., Wong, C. K. H., Wu, T., Chen, D., Lu, C., Chen, W., & Guo, V. Y.
(2022). Comorbid depression and obesity, and its transition on the risk of functional
disability among middle-aged and older Chinese: a cohort study. BMC Geriatrics, 22(1),

https://doi.org/10.1186/s12877-022-02972-1
Micco, J. A., Henin, A., Mick, E., Kim, S., Hopkins, C. A., Biederman, J., & Hirshfeld-Becker,
D. R. (2009). Anxiety and depressive disorders in offspring at high risk for anxiety: A
meta-analysis. Journal of anxiety disorders, 23(8), 1158-1164.
Nazabal, A., Olmos, P. M., Ghahramani, Z., & Valera, I. (2020). Handling incomplete
heterogeneous data using vaes. Pattern Recognition, 107, 107501.
Shieh, G. (2019). Effect size, statistical power, and sample size for assessing interactions
between categorical and continuous variables. British Journal of Mathematical and
Statistical Psychology, 72(1), 136-154.