Upgrade to remove ads
Only $35.99/year

Inman; Domain I Topic C

Terms in this set (60)

descriptive research
describes nature at a point in time
-generate hypotheses regarding determinants of a condition or disease.
-does not prove Cause/Effect.

qualitative research
case studies, case reports, case series
surveys
qualitative research
purpose is to explore a phenomena of interest as a prelude to theory development.
-what specific dimensions are assoc with being an effective counselor?

collect data through focus group, delphi, interviews, questionnaires, observations.
case studies
describes quantitatively the experiences of a group of cases with a disease or condition in common.

help to determine important etiology, care, or outcomes of a particular condition.
surveys
research designed to describe and measure characteristics of a defined population
-defined time frame
-pinpoints the problem.
analytical research
tests hypotheses concerning the effects of specific factors of interest.
-allows casual association to be determined:
can prove cause/effect.

experimental
quasi-experimental (time series)
cohort studies
case controls studies
cross-sectional studies
experimental studies
study in which a treatment, procedure, and program are intentionally introduced to monitor an outcome.
-randomization used: randomly chosen to be in control/placebo group.
-Hawthorne effect

*determine the difference btw two groups, if one group has improved more than the control it was successful.
Quasi-Experimental Design
time series: series of measurements at periodic intervals (before/after) program.
-continuation of previous patterns or whether they results indicate a noteworthy change.

*no control group
use of pre/post test.
Cohort Studies
-other name for this study?
-3 types of cohort studies
any group of people who have something in common, followed through time to see if they develop a specific disease.
"incidence studies"= frequency of new cases.

-carried out over long period of time (longitudinal)
-prospective (future-oriented)
-studies existing data and look back at relationship btw exposure factors and outcomes (retrospective)
Case Control Studies
focus on specific disease.
-compare those with the disease vs. those without the disease (but otherwise similar characteristics).

*both groups recall past behaviors, to study how the groups differ.
Cross-sectional Studies
one time data collection.
"prevalence study": ALL of the cases of a specific disease amount a group of people in a specific time.

-snapshot at one point in time; describes CURRENT SITUATION (does not look in the past or the future).
relevance/ validity
is the ability to measure phenomenon it intends to measure.
internal validity
tests whether difference btw 2 groups is real.

-how well was the exp done? how good was it at avoiding confounding variables?
external validity
* to which, generalization can be made from the study to the larger population.
what is an analysis of variance?

what does it ask?
-what does it answer?
-when is it used?
ANOVA
tool used to evaluate validity of more than 2 samples!.

-asks whether the difference btw samples is reliable one that would be repeated.
-used when more than 2 are compared against each other.
-are there one or more significant differences anywhere among the samples?!
reliability
consistency or reproducibility of test results.
-test, then retest later. are the results the same?
precision
-less ____ means greater _____
amount of VARIATION that occurs RANDOMLY.
-LESS random VARIATION= GREATER PRECISION in the measurement and greater reliability.
sensitivity vs. specificity
-when are these two used?
-define each
used if protocol involves SCREENING for a PARTICULAR condition.
-evaluates the cut-off value being used.

specificity: proportion of non-afflicted ID as non-afflicted.
sensitivity: proportion of individuals who test positive.
variables


-4 types
characteristics that have different values from observation to observation

-nominal (non ordered)
-rank order ( ordinal scale)
-numerical discrete
-numerical continuous
nominal (non ordered)
gender, race, marital status, present or absent.

variables that fit into categories.
rank order (ordinal scale)
observations compared with each other and put in order.

-state of disease from 1 to 4
-order from best to worst.
numerical discrete
data with numbers
(# of clinic visits)
numerical continuous
underlying continuous scale
BP, HR, wt, exam score
sampling: probability


-randomization
each segment of the population will be represented in the sample.

-randomization: select from sample from whole population so the characteristics of each of the units approximate the characteristics of the whole population.
samples: non-probability



-2 types of non probability sampling
no way of forecasting that each element in the population will be represented in the sample

-convenience
-quota
convenience sampling
take units as they arrive on the seen, who ever is around is part of the study.
-no attempt to control bias.
quota sampling
select units in the same ratio as they are found in the general population
ie: 40% females 60% males in state population
total sample should be divided so 40% are female, 60% are male.
measures of central tendency
-examples
center of any mass of data:
-mean (average)
-median (mid point)
-mode (most frequently occurring value)
measures of dispersion
-examples
how values are distributed about the mean
-range
-standard deviation
range
difference between the lowest and highest value in the distribution.
standard deviation

-pont of inflection; where is it located on a normal bell curve?
-where is 1 SD found?
-__%= observations lie in 1SD of the mean
-__%= observations lie in 2SD of mean
-__%= lie outside the range.
indicates degree of dispersion about the mean value.
-point of inflection: when bell starts to go inland slope begins to level off.
distance between mean and point of infection on either side= 1 SD.
*about 68% of all observations in a normal distribution lie within 1 SB of the mean.
*32% observations lie OUTSIDE 1 SD (16 below -1 SD/16 above +1 SD. )
*95% lie within ~2SD either side of the mean.
existence of relationships:
-degree of linear relationship

-linear correlation coefficient:
value is always between __ and ___
the closer the pts are in the line, the stronger the degree of linear relationship.

-(r)
always btwn -1 and +1
r=+1
slope?
all pts lie exactly on a straight line with a positive slope.
r= -1
slope?
all points lie on a straight line with a negative slope.
r=0
no linear relationship
-the closer r is to 0, the more dispersed the points will be from the line.
negative correlation vs. positive correlation?
negative (inverse relationship)
- increase in HDL, decrease in Heart Disease Risk

positive (same direction)
-increase in LDL, increase in HD risk.
line graphs
frequency of y axis
method on x axis
bar charts
-show measurement only on vertical axis
-bars arranged horizontally or vertically in ascending or descending order.
histogram
-what is it? how is it set up?




-what does it summarize?
block diagram, who's blocks are proportional in area to the fq in each class or group.

fq on Y axis
categories on X axis.

***summarize data from a process that has been collected over time
double blind study
removes bias from research
-neither researcher nor subjects know if they are in the tx group or control group.
mortality
rate of death
morbidity
state of disease
variable of interest
what researchers are observing
population of interest
describes group about which the observations are made.
descriptive statistics
summarizes and describes aspects of a set of data.
inferential statistics
technique that allow conclusions to extend beyond an immediate data set.

-what is the probability that the results can be applied to a larger group?
-what can you infer from these results of your study?!
non parametric test
does not depend on a normal distribution

"skewed curve:
data are counted or ranked
-discrete/categorical
-mode is peak of frequency
*non-parametric does not take into account the average or variance (distance from the mean)
*nominal and ordinal"
dichotomous scores
only two events are possible
ie: heads/ tails
continous scores
measures on a continuous scale.
pilot study
-purpose
-what are you determining?
scaled down version of the larger investigation
-practice implementation
-determine whether clinical trial, as planned, is feasible?
-are goals realistic and attainable?

*includes every step in the study, but done on a scaled down group to see if the study plan works!
focus group

-what do these groups talk about?
-what type of data is this?
method of attaining info about a target group.
sm. group talks about beliefs, opinions, problems.

*contributes to attitudinal data.
-subjective information.
chi square X
tests whether there is a true difference between categories.

-attributes that have more than 2 categories.
-compares fq. with which we'd expect certain observations to occur w/ fq that actually occured.
t-tests
tests significance betwn the means of two different populations.

-tests null vs. alternative hypothesis.
-if probability value is equal to less than the level set for significance the null is rejected.
ie: if it is <.05 then we can reject the null.
7 steps in research
1. ID a relevant topic, review published research literature r/t topic.
2. formulate research Question
-who, what, how?
3. research question leads to hypothesis
4. prepare research protocol
-methodology to solve problem.
5. contain and organize materials/ methods.
6. collect and analyze data
7. study results and make decisions.
8 parts of a research report (in order)
1. abstract
2. general introduction
3. review of existing literature
4. methodology
5. results
6. discussion
7. conclusion
8. implications
abstract
condensation of final report
-purpose
-questions asked
-scope and method
-summary of conclusions
results
specific labs
clinical findings
objective findings
subjective findings
discussion
interpretation of results
-comparison with other studies.
-may be combined with results.
conclusion
brief summary of results
-may have recommendations
interpretations
how info might be applied in practice
-how they'll use the found results.
hypothesis vs. null hypothesis
-hypothesis:
prediction of a relationship

-null hypothesis:
no relationship in population of data.
-any difference is result of sampling error
"equal to"
Sets with similar terms