research process + methods Flashcards
Types of research
pure:
- Concerned with generating new theories and knowledge for its own sake
-ex: Theory development
experimental:
- manipulation of one variable to see its effect on another, while controlling for as many other variables as possible, and randomly assigning subjects to groups
- ex: Double-blind random assignment control groups, response to an intervention
clinical:
- Conducted in a clinical setting where control over variables is difficult.
- Example: Drug trials, therapeutic results
applied:
- Aims to answer practical questions that help people perform their jobs better
- practical and focused on creating immediate benefits or improvements for real-world problems
-ex: time use studies, evaluation of interventions with the same purpose
descriptive:
- Focuses on describing a group, situation, or individual to gain knowledge that may be applied to other groups or situations, often used in case studies or trend analyses.
- Example: Surveys, qualitative research, measurement of characteristics, response to phenomena
laboratory:
- performed in lab setting
- basic science research
Types of research examples
pure:
-Theory development
experimental:
- Double-blind random assignment control group
- response to an intervention
clinical:
- Drug trials
- therapeutic results
applied:
- time use studies
- evaluation of interventions with the same purpose
descriptive:
- Surveys
- qualitative research
- measurement of characteristics
- response to phenomena
laboratory:
- performed in lab setting: basic science research
Developing research project
*Choosing project and developing design depends on:
- Knowing what is expected
- Identifying clear parameters
- Following organizational guidelines
*Many organizations prescribe scientific writing format
*Institutions may assign topics and specific designs
*Depends on many factors:
- Time, effort, cost, resources, and ability
- Personal attributes, interests, resources, and expectations of self
- Strengths as researcher and ability to accomplish project
*Seek out people with skills in area of interest and ask for their help
Null hypothesis and rejecting it + hypotheis
Null Hypothesis (H0- There is no connection between X and Y)
- researcher conducts his research and if she finds that there IS A CONNECTION (ie-X does cause Y), then the Null Hypothesis is rejected
hypothesis: Ha or H1:
- x causes y
- researchers look for proof that this is true
-ex: commutes over 60 min leads to work dissatisfaction
4 Components for Measuring Disease Frequency
- population: which groups
- cases of disease : numerator
- size of population: denominator
- time: be explicit
Population definition
-defines the specific group in which the disease frequency is being measured
- should be well-defined and can be based on characteristics such as age, gender, geographic location, occupation, or other relevant attributes
- simple definition: group of people with a common characteristic
Can be any common characteristic:
*Residents near a hazardous waste site
*Workers at a nuclear power plant
*Diners at a restaurant
*Students who use health clinic at New York Tech
Population: fixed vs dynamic
Fixed:
- population with membership by defined EVENT (PERMANENT MEMBERSHIP)
- ex: veterans in vietnam war, people born in 1982
dynamic:
- population with membership defined by STATE OR CONDITION (TRANSIENT MEMBERSHIP)
- ex: residents of boston, parents of a teenage, # undergrads
Cases of diseases definition
Disease:
- ANY health outcome (infection, defect, injury
Cases of disease:
- number of new or existing cases of ds within a defined population
- NUMERATOR
- attain cases with clinical records, dx tests, disease registeries, surveillance, self reporty -> QUALITY of data is key
Size of population: definition
DENOMINATOR for all measure
- size of population: based on population you identified
- can be full or sample of population
- necessary to compare diseases across population (cannot just use #cases)
Time: measure of diseases frequency importance
Time = necessary for all measures of disease frequency
Disease occurrence can be measured at single point in time
- At study enrollment
- at birth
- on a particular date
Disease occurrence can be measured over an INTERVAL of time:
- example: study on autism spectrum disorders, for a period of follow-up from birth to 10 years of age, while living in the town of Old Westbury
Three generic types of measure: ratio, rate, proportion
Ratio:
- divide one number by another (numbers do NOT need to be related
- ex ratio: observed #AIDS cases in country A/expected #AIDS cases in country A = 2 (2x as many cases as expected)
Proportion:
- division of two RELATED NUMBERS
- numerator is a subset of denominator (often a percentage)
- ex: If 1000 blood samples were collected and 120 were positive for Covid, the proportion of positives is 120/1000 or 12%. It is said that 12% of intensive care patients were infected with Covid
Rate:
- specific type of ratio: TIME IS THE INTRINSIC DENOMINATOR
what are the three main measures of disease frequency? what characteristics do they share
- prevalence
- cumulative incidence
- incidence rate
all share characteristic:
- numerator = number of cases
- denominator = size of population
- measure of time
- these measurements are made for specific source population
prevalence + point prevalence
Prevalence:
- measures how many people have a disease at a certain time (#EXISTING CASES)
- It’s like taking a snapshot of a disease in a population at one specific point in time or over a specified period
- involves BEING IN A STATE: someone has the ds at the time you are measuring it (snapshot) -> pt could recover later
- denominator: total population
Point prevalence:
- is the number of existing cases at a SINGLE POINT IN TIME divided by the total number of individuals in the population at that same time
- ex: on july first, at birth
-
What is the prevalence of arthritis in City A on January 1, 2017?
*City A has 7,000 people with arthritis on January 1, 2017.
*The total population of City A is 70,000 on that same day.
incidence:
Definition:
- measures occurrence of NEW CASES of a disease in a population during specified period
- usually first time occurrences of new ds
- denominator: ONLY POPULATION AT RISK * -> excludes people who are immune, people who already have the disease
- involves the TRANSITION from one state to the other
ex: incidence of death: alive -> death
prevalence vs incidence
Main thing:
- prevalence: existing ds
- incidence: NEW disease
prevalence:
- It is used to assess the BURDEN OF DISEASE and is important for the administration, planning, and allocation of resources.
- It is used in RESEARCH when INCIDENCE IS DIFFICULT TO MEASURE -> ex: conditions with an uncertain onset (like MS or depression)
incidence:
- Important for ETHIOLOGICAL RESEARCH to understand what is causing the disease
- Useful for evaluating the EFFECTIVENESS OF PREVENTATIVE MEASURES and whether they work in preventing new disease
-Helps in the EVALUATION OF TREATMENTS to see if they improve survival and quality of life for those affected by the disease
crude mortality rate
total number of deaths from ALL CAUSES within a population over a specific time period, often per 100,000 people per year
cause specific mortality rate + age specific mortality rate
Cause-Specific Mortality Rate:
- total number of deaths from a specific cause (like AIDS) within a population over a certain time frame
- per 100,000 people per year
Age-Specific Mortality Rate:
- total number of deaths from ALL CAUSES among people within a SPECIFIC AGE category (e.g., less than one year old) in a population over a year, reported per 100,000 people.
Special types of cumulative incidence rates: 2 types
Attack Rate:
- measures the proportion of individuals exposed to an infectious agent who become infected over a certain time period
- number of new cases of a disease divided by the population at risk at the beginning of the time period
Case Fatality Rate:
- proportion of individuals with a certain disease who die from that disease within a specified time period.
- number of deaths due to a specific disease divided by the number of people who have the disease.
sample size and statistical power
sample size:
- number of participants in the study, including both the group receiving the intervention (if applicable) and the control group
Statistical power:
- refers to the likelihood that the study will detect an effect or difference if there is one
- A study with high statistical power has a better chance of detecting a true effect
- is the number of sample size adequate to demonstrate statistical significance if the study hypothesis is true?
- To ensure sufficient statistical power, researchers calculate the necessary sample size before starting the study based on the expected effect size, the variability of the data, and the desired level of statistical significance
MAARIE Framework for studying a study
Method: design of study
Assignment:
- involves selecting a study population and a comparable control population
- ex: incidence rate of strokes in young women who use a particular type of birth control compared to those who don’t
Assessment:
- process of collecting the data you are interested in
- determine the prevalence of use of birth control pills in the study population and the control group
Results:
- see if there’s a notable difference or association with study and control group
Interpretation:
- interpret results for individuals IN YOUR STUDY
- risk of stroke among women in YOUR STUDY
Extrapolation:
- extend your conclusions to a broader context
- ex: draw conclusions for women NOT IN YOUR STUDY like ones that have the option to use newer low-dose birth control pills
Case control study vs cohort study vs randomized controlled trial in context of birth control and stroke risk
case control:
- Assignment: Researchers select two groups: one that has the condition (cases) and one that does not (controls)
- select group of women who have had a stroke and group of women who have not had a stroke ->
-OBSERVED ASSIGNMENT
- assignment is observed rather than manipulated
cohort:
- Assignment: A group using birth control pills (the exposed cohort) is compared to a group not using them (the unexposed cohort)
- without the investigator’s intervention
randomized:
- Assignment: Women are randomly assigned to either receive birth control pills or be in a control group, ensuring the groups are comparable
- with investigator intervention
