Research Flashcards
Describe dependent variable
AKA the outcome variable is the measure in which you are interested as the outcome of researcher i.e. Study factors influenceing exclusive breast-feeding ( breastfeeding is the dependent variable)
What is the independent variable
AKA the explanatory variable - measures used to explain the variaation in the outcome ( ie maternal age, counseling, birth weight etc)
NO relationship between dependent and independent variables is called ……….
A null hypothesis ( researcher must find the proof)
What is the P value
the probability of seeing your research result based on chance alone.
Explain P<.05
If the null hypothesis is correct you would see a result like this BY CHANCE less then 5% of the time Meaning the results were not by chance and that the research was statistically significant.
Interpret P<.001
P<..001 less then .1 % chance
all are statistically significant.
How would you determine that research was NOT statistically significant
Any study that has a P value greater then 5%
P<.20 and you would then conclude a null hypothesis– there is no statistically significant differnce or relationship
What is the power of a study?
This refers to how likely you will find a real diffence if it exists given your sample size. Power should be 80% or more
Clinical significance ?
If the study has a clinical impact ( often seen in a large sample size resulting in only a small diffence due to large numbers but would have implact clinically)
what is Odds ratio
the odds of getting a disease in an exposed group compared to unexposed group ( # of times it happened divided by # of times did NOT happen)
For odds ratio or RR (relative risk ratio) how do you measure significance
1 means there is the same risk, >1 =bigger risk and <1 smaller risk. Often measured with 95% CI (confidence interval)
Explain confidence interval (CI)
if CI includes 1 then NS ( not statistically significant) (ie
If 95% CI 0.9-1.5 this would be NS because 1 is included in the range
95%CI 1.3-1.5 does NOT include “1” so therefor this would be statistically significant.
Determine significance if RR 1.4 (95%CI 1.2-1.6)
This is statisically significant because the
RR is >1 and the CI does not include 1
Describe relative risk
Risk ration comparing 2 groups - what is the risk for diseae for the exposed group compared to the unexposed group
What is difference between cross sectional study or longitudinal study
cross sectional - one point in time
longitudinal - over a period of time
What is external validity
If the results of the study are generalizable to the settings beyond the study itself
Survey instruments measure from various sources- what are two criteria to determine effectiveness
Reliability and validity
What is a “type 2 “ error in a study
When a study concludes that there is not difference ( null hypotheses is correct) even though there really was a difference at the population levels
two types of research design
descritptive and analytical
if a research design is analytical and there is artifical manipulation involved then this is call
experimentation
If women have sore nipples and are randomly assigned two types of nipple cream and are followed to see which is the fastest healing- what type of study woul this be
randomized control study
2 similar hospital sites are chosen and one begins a new rooming in policy intervetnion while the other does not. In hospital breastfeeding rates are measured to see if the policy had a effect. choosing tow similar sites makes this __________
a quasi experimental comparison group
Information collected concurrently at ONE point in time refers to
a cross sectional study. the results of a cross sectional study implies association of the factors NOT one factor causing the other factor. ie low income associated with low breast feeding rates studied at one point in time
Information collected over a time period is
a longitudinal study
