Topic 1: Principles of Research
Quantitative Data: Researchers should get the same numbers from the
data.
Numeric – measurable, graphs, tables, etc.
Verifiable and/or accurate – value of the data is in the number.
Examples: Sample size (n), mean, median, mode, range, standard
deviation.
Qualitative Data: It differs. It is subjective.
Observed, categorical – the value is in the category.
Examples: Pain scale, symptoms, interviews of staff, etc.
P-value: The probability of chance. What is the probability or what is the
chance that what is going to be observed was going to happen anyway..?
Variables:
Independent Variable: has an effect on the dependent variable. It is
what you are controlling.
Dependent Variable: is what is going to change (or not) as a result
of research or intervention. (Example: you are researching whether a
new exercise machine has an effect of lowering B/P. The independent
variable is the exercise machine, and the B/P is the dependent
variable.)
Confounding Variable: obscures the effect of another variable.
(Example: From the above scenario, you find out that four research
participants did not disclose that they are taking blood pressure
medications.)
Ordinal Variable: order matters, but not the difference in value.
(Example: Birth order.) Placement is key!
Continuous Variable: aka – Interval variable – meaningful distance
between values. (Example: Temperature scales.)
Dichotomous Variable: aka – Binary Variable – It occurs in two
possible states. (Example: A patient is diabetic or non-diabetic.)
Categorical variable: assumes values that are names or labels.
(Examples: Eye color, Breed of Dog.)
Research Studies:
Cross-sectional: a type of study that analyzes data collected from a
population or subset at a specific point in time. The amount of data
, pulled is pre-determined and can be as big or small as warranted.
(Example: cooking with an onion; once the onion is cut, it cannot be
uncut.) After 6 months, the results of the study will be analyzed. Only
one specific point in time.)
Time-Series: method for analyzing a sequence of data taken at pre-
determined points in time. Does not always need to be “neat.” We use
multiple cross-sectional data pulls to achieve our analysis (more than
two data pulls becomes a time-series.) (Examples: weekly for 12
weeks, for 3 years, or Weeks 1,7,10,13,18,22.)
Focus Group: a demographically diverse group of people assembled
to participate in a guided discussion about a particular product or
process. (Example: A healthcare organization wants to ask its nurses
their opinion on a new staffing policy.) Qualitative in nature!
Case Study: researchers investigate one person, one group, or one
institution in depth. Most often seen in professional journals and often
focuses on a particular problem and how it was solved. (Example: in
a future course you will study Sentara Health and how it managed to
become on of the first integrated delivery systems.) Can use both
methods of research; interviews = qualitative, focus groups, reports,
or data analysis. It is a mixed method!
Cohort Study: a study normally used to investigate the causes of
disease or establishing links between risk factors and health
outcomes. Forward looking, Prospective study, Planned in advance
and carried over a future period of time (longitudinal study).
(Example: Researchers want to follow 5000 individuals with a family
history of dementia. They may send annual surveys to track lifestyle
(diet, exercise, medications), work history, and social history over a
period of 20 years to determine the effect that genetics has in
developing the disease or if certain lifestyle choices helped mitigate
development.)
Literature Review: Learning for learning’s sake. (Example: Reading
about AI and uses in healthcare.)
Meta-analysis: think of this as a literature review with action.
(Example: Reading about AI in healthcare to create a new diagnostic
tool.) Literature Review + Verb.
Factor Analysis: a process in which the values of observed data are
expressed as functions of a number of possible causes in order to find
which are the most important. (Example: I want to rank the reported
reasons of nursing turnover in terms of frequency.)
Quantitative Data: Researchers should get the same numbers from the
data.
Numeric – measurable, graphs, tables, etc.
Verifiable and/or accurate – value of the data is in the number.
Examples: Sample size (n), mean, median, mode, range, standard
deviation.
Qualitative Data: It differs. It is subjective.
Observed, categorical – the value is in the category.
Examples: Pain scale, symptoms, interviews of staff, etc.
P-value: The probability of chance. What is the probability or what is the
chance that what is going to be observed was going to happen anyway..?
Variables:
Independent Variable: has an effect on the dependent variable. It is
what you are controlling.
Dependent Variable: is what is going to change (or not) as a result
of research or intervention. (Example: you are researching whether a
new exercise machine has an effect of lowering B/P. The independent
variable is the exercise machine, and the B/P is the dependent
variable.)
Confounding Variable: obscures the effect of another variable.
(Example: From the above scenario, you find out that four research
participants did not disclose that they are taking blood pressure
medications.)
Ordinal Variable: order matters, but not the difference in value.
(Example: Birth order.) Placement is key!
Continuous Variable: aka – Interval variable – meaningful distance
between values. (Example: Temperature scales.)
Dichotomous Variable: aka – Binary Variable – It occurs in two
possible states. (Example: A patient is diabetic or non-diabetic.)
Categorical variable: assumes values that are names or labels.
(Examples: Eye color, Breed of Dog.)
Research Studies:
Cross-sectional: a type of study that analyzes data collected from a
population or subset at a specific point in time. The amount of data
, pulled is pre-determined and can be as big or small as warranted.
(Example: cooking with an onion; once the onion is cut, it cannot be
uncut.) After 6 months, the results of the study will be analyzed. Only
one specific point in time.)
Time-Series: method for analyzing a sequence of data taken at pre-
determined points in time. Does not always need to be “neat.” We use
multiple cross-sectional data pulls to achieve our analysis (more than
two data pulls becomes a time-series.) (Examples: weekly for 12
weeks, for 3 years, or Weeks 1,7,10,13,18,22.)
Focus Group: a demographically diverse group of people assembled
to participate in a guided discussion about a particular product or
process. (Example: A healthcare organization wants to ask its nurses
their opinion on a new staffing policy.) Qualitative in nature!
Case Study: researchers investigate one person, one group, or one
institution in depth. Most often seen in professional journals and often
focuses on a particular problem and how it was solved. (Example: in
a future course you will study Sentara Health and how it managed to
become on of the first integrated delivery systems.) Can use both
methods of research; interviews = qualitative, focus groups, reports,
or data analysis. It is a mixed method!
Cohort Study: a study normally used to investigate the causes of
disease or establishing links between risk factors and health
outcomes. Forward looking, Prospective study, Planned in advance
and carried over a future period of time (longitudinal study).
(Example: Researchers want to follow 5000 individuals with a family
history of dementia. They may send annual surveys to track lifestyle
(diet, exercise, medications), work history, and social history over a
period of 20 years to determine the effect that genetics has in
developing the disease or if certain lifestyle choices helped mitigate
development.)
Literature Review: Learning for learning’s sake. (Example: Reading
about AI and uses in healthcare.)
Meta-analysis: think of this as a literature review with action.
(Example: Reading about AI in healthcare to create a new diagnostic
tool.) Literature Review + Verb.
Factor Analysis: a process in which the values of observed data are
expressed as functions of a number of possible causes in order to find
which are the most important. (Example: I want to rank the reported
reasons of nursing turnover in terms of frequency.)
