100
what is research?
Systematic inquiry that uses disciplined methods to answer questions and solve problems
what are the type of Hypotheses & example
Null(H0)- "No effect' (eg no significance in flossing)
Alternative(Ha) - "effect exist" (eg: significance seen in flossing)
example:
research question:Does dental flossing reduce the number of cavities?
how to get Qualitative Trustworthiness
Credibility: Triangulation, member checks.
Transferability: Thick descriptions.
Dependability: Audit trail.
Confirmability: Reflexivity.
MNEMONIC: "Can This Dog Catch?" (Credibility, Transferability, Dependability, Confirmability).
what is research design?
the overall plan/strategy that guides a research project from its conception to the final analysis of data. (like a blueprint as a researcher how you might find & analysase data while ensuring consistency, reliability and validity)
what are the qualitative research design methods.
Descriptive qualitative (general description - just observation and interviews no theory)
Narrative inquiry (personal stories)
Grounded theory
Phenomenology
Ethnography
what is the focus of nursing research
develop & refine- knowledge thus improving clinical practice and care outcomes
Generate- generate new knowledgeto improve clinical outcome & care outcomes
Build- build on evidence based practice in nursing practice
RMB DR GB
type of error
Type I Error (False Positive):
You think your new drug works, but it doesn’t.
Mnemonic: "Type I = False Alarm" (like a fire drill for nothing).
Type II Error (False Negative):
You miss a real effect (e.g., drug works but you dismiss it).
Mnemonic: "Type II = Missed Opportunity".
non probability sampling (Qualitative)
got 4 type non probability
Purposive sampling (judgement/subjective sampling)
limitation: bias
Convenience sampling: participants based on their availibility/accessibility
- strength: quick and easy
- limitation: researcher bias
Snowball sampling (relies based on referrals based from initial participants to recuits additional participants) (used to get into reaching hard to access population
- strength: un
-limitation: bias, unable to produce a representative sample
Maximum variation sampling- widest possible range of perspectives, experiences, or characteristics related to your study topic.
-Diversity-Driven, Not Random:
eg: Young vs. elderly patients.
(Aims for deep insights, not statistical representation , Ideally obtain multiple perspectives on issue - Sample size often linked to data saturation )
which to choose need to be guided by: research question, aims,
what are the quantative research design methods.
Descriptive
correclational
experimental
quasi-experimental
phenomonological research
using indepth interview
open ended questionaires
Strrength: deep emotion
limitation: small sample size, researchers bias
deductive reasoning v inductive reasoning
deductive:
-developing specific predictions from general principle
-starts general theory/hypothesis then test thru specific observation and experiments to confirm initial theory
-to test hypothesis/assumption
inductive:
- reasoning from specific observation to assumptions
- starts with observation, experiment and measurement, generalised finding pattern in data.
- to develop hypothesis/assumption
what are the different variables?
IV (Independent): What you change (e.g., drug dose). (control group)
DV (Dependent): What you measure (e.g., pain level). (experimental group)
data collection methods (IDDAF)
Interviews/ Focus Group Interviews
Direct Observation
Documents
Audio visual materials
Field notes
Triangulation – utilize various data sources to describe the phenomena
Descrptive research design
strength: purely describe the 5w1h, precursor for other design
grounded theory
aims to develop theories by continously analysing and comparing collected data from relatively large number or participants
(inductive approach- bottom up approach)
collecting data through interviews or observation
then analyzing it to identify patterns and themes
and keep collecting until data saturation (meaning no need information -nothing to squeeze alr)
strength: theory generation (under research area)
Limitation: time intensive
Quantitative v qualitative
Quantitative
- Numbers, stats, hypotheses
-Deductive (test theory)
-RCTs, surveys ->Example: Pain scale (0-10)
Qualitative
- Words, themes, experiences
Inductive (build theory)
Interviews, observations ->Example: Patient stories about pain
MNEMONIC: "Quant = Numbers, Qual = Stories"
what are the levels of measurements? (BNOIR)
Binary variable: Categories with only 2 option (e.g., gender, Do you want pepperoni?" → Yes or No. ).
Nominal: Categories with more than 2 option (e.g., vegetatrian).
Ordinal: Ordered (e.g., pain: mild/moderate/severe).- got ranking
continuous (Are the one below) - has equal interval
Interval: No true zero (e.g., temperature).
Numbers where the distance between values is meaningful, but zero doesn’t mean "nothing".
Like a ruler with no true starting point.
Ratio: True zero (e.g., weight).
Numbers where distances AND ratios matter, and zero means "none".
MNEMONIC: "BNOIR" (binaryNominal, Ordinal, Interval, Ratio).
What’s triangulation?
Using multiple data sources (e.g., interviews + observations) to validate findings.
correlational research design
involves identifying and measuring relationship between 2 or more variables without manipulating them
good to use if you want to know potential relationship that cant be manipulation of variables
strength: large scale
limitation: casualty (correlation does not equal causation) (causation use experimentation)
ethnographic research design
invloves observing and studying a culture sharing group of people in their natural setting to gain insights into their behaviour beliefs and value
- observation
- in-depth interviews to explore participants view, beliefs etc
strength: generates rick, contectually embedded insights into the socio-cultural dynamics of human behaviours
limitation: researchers bias, participants confidentiality, ethical complexities
probability sampling (mostly use in quantitative study)
- selecting participants on a statistically random basis
simple random sampling (randomly pulling numbers out (using random number)
Strength: easy to implement and cheap
Limitation: lack of minority representation in the subgroups
Statified random sampling: selecting randomly but from within a certain predefined subgroups that share a common trait (eg: gender, level of education etc) (heterogenous- more mix/varied)
strength: more control over the impact of large subgroup within the population
more precise as well
cluster sampling: sampling from naturally occurring mutually exclusive clusters within a population (eg: areacode: cus neighood ppl might share same income eg landed ppl usually earn more) (homogenous- more similar )
diff between experimental and nonexperimental
1) experimental research
Researchers actively introduce an intervention or treatment, most often to address Therapy questions.
Classified as clinical trials in medical research
Test causal relationships
2) nonexperimental research
collect data without intervening or introducing treatments
•Classified as observational studies in medical research
quasi experimental resarch design
used to investigate casual relationshi[ when researcher cannot/dw to randomly assign participants
(uses pre-existing issue to do grouping)
Strength: larger scale -> means more statical power
Limitation: more difficult to confidently establish causality, less control -> might cause more confounding variable
experimental research design
used to determine if there's casual relationship between variables (independent variable (control) and dependent variable)
strength:identify and measure casual relationship
limitation: challenging to control all variable hence small sample size
random assignment (reduce bias -> ethical issue maybe)
what are the non experimental research designs?
Non-Experimental
Cohort: Follow exposed vs. unexposed over time (prospective).
Case-Control: Compare sick vs. healthy (retrospective).
Cross-sectional: Snapshot (e.g., survey on supplement use).
MNEMONIC: "RCT = Random, Cohort = Future, Case-Control = Past"