IR WEEK 1 Flashcards
describe ways of knowing
scientific method, deductive reasoning, inductive reasoning, trial and error, authority, tradition
includes deduction and induction in systematic empirical and controlled analysis
scientific method
gather facts and observations without preconceived notions, facts about a sample leads to conclusions about about a whole
inductive reasoning
start with a premise or known scientific principle, tie that premise in with other observations and make a conclusion
deductive reasoning
keep trying until something works
trial and error
turn to an expert
authority
we accept certain truths as givens
tradition
3 major parts of the research continuum
Descriptive, exploratory, experimental
the researcher attempts to describe a group of individuals on a set of variables, to document their characteristics
descriptive
researcher examines a phenomenon of interest and explores how it relates to other factors (find relationships)
exploratory
compares two or more conditions (determines cause and effect)
experimental
5 major phases of the research process
- identify the research question
- design the study
- carry out methods
- data analysis
- communication
enough evidence accumulates to discredit an existing theory; may cause a change in perspective
paradigm shift
List and describe the 7 parts of a primary research article
Abstract, Introduction, Methods, Results, Discussion, Acknowledgements, References
overview and purpose
abstract
statement of the problem; specific purpose and hypothesis
introduction
study design; data analysis and procedures
methods
narrative description of statistical outcomes
results
interpretation of statistical outcomes
discussion and conclusions
technical assistant; funding source
acknowledgments
all references cited in articles
references
synthesizes the data from multiple research studies and provides an argument or interpretation of the state of the field
review paper
data sets are split and published separately instead of being presented in unified way
salami science
a characteristic that can be manipulated or observed an that can take on different values, either quantitatively or qualitatively
variable
variables that might change in response to some intervention
dependent
condition, intervention or characteristic that will predict or cause a given outcome
independent variable
declarative statement that predicts relationships between independent and dependent variables in a specific population
hypothesis
alternative/research hypothesis
best guess answer to the research question; the expectation that is to be tested
null hypothesis
opposite of the research hypothesis; challenging the alternative hypothesis.
non directional hypothesis
no indication of change, higher or lower
directional hypothesis
predicting that something is either going to go higher, lower, or stay the same
the larger group to which research results are generalized
population
researcher chooses a subgroup of the population; reference group for estimating characteristics or drawing conclusions about the population
sample
individuals selected for a sample over-represent or under-represent certain population attributes that relate to the study variables
sampling bias
factors that preclude or prevent someone from being a subject
those with conditions/ diseases additional to the disease under study
extreme ages
certain disabilities or deficits that can affect how they can participate
factors of inclusion; population that qualifies someone as a subjec
gender
disease state having particular treatment
only certain symptoms
every unit has equal chance of having some of the characteristics throughout the population. The sample should be representative of the population
random sampling (probability sampling)
groups should be equivalent. differences between groups have been distributed as a function of chance alone
random assignment
the difference between average (statistics) and population averages (parameters); should be due to chance, not intentional error or bias
sampling error
experiment requirement through which change is observed. If we observe a change in the treatment group but not the control, we can attribute the change to the treatment.
control group
factors that contaminate the independent variable so that separate effects of the variable are obscured
extraneous variables
not measured in the study but can have confounding effects
lurkers
either the subjects are unaware of treatment type but the researcher knows or vice versa
single blinding
neither subjects or investigators know the identity of treatment groups until after data are collected or analyzed
double blinding
List the three essential characteristics of an experiment
- Independent variable manipulated by researcher
- A control group must be incorporated into the design
- Subjects must be randomly assigned to groups
participants knowledge of treatment status or the investigators expectations can influence performance or reporting of outcomes
experimental bias
every person in the population has an equal chance of being chosen; sample should be representative of the population
random sampling
convenience sampling chosen on basis of availability. Ex: asking for volunteers by posting signs or advertisements
non-probability sampling
Explain why statistics are important in research
Application of mathematical methods to research conclusions about the world made from observations (experiments)
Allows us to make accurate inferences from our incomplete observations
an educated guess, a process of inferring features of a population by looking at a small sample
Address the questions of the likelihood that an observed difference between two groups could have arisen by chance
inferential statistics
describe data only (does not manipulate or compare data)
Measures of central tendency
Measures of variability
descriptive statistics
numbers that tend to cluster around the middle of a set of values
measures of central tendency
the score that occurs most frequently in a distribution
mode
the value above which there are as many scores as below it. Divides a rank-order distribution into 2 equal halves
median
(X-bar) sum of a set of numbers divided by the number of scores, n
mean
an observation that falls well above or well below the overall bulk of the data
outliers
difference between largest value and the smallest value. Only uses the largest and smallest observations ignoring the rest of the data
range
the typical value of how far the data falls from the mean, by summarizing the deviations from the mean
standard deviation
Statistical inference is an educated guess, a process of inferring features of a population by looking at a small sample; addresses questions of likelihood
inferential statistics
choosing of samples as representatives of the entire population
sampling
the distribution obtained by computing the statistic for a large number of samples drawn from the same population
sample distribution
FALSE POSITIVE when we conclude that a real difference exists, when the differences are in fact due to chance.
type 1 error (alpha)
FALSE NEGATIVE when we conclude that the differences are due to chance when the samples are truly different.
calling the results not statisitically significant when they really are
type 2 error (beta)
consists of arbitrary labels with no implied order; unranked; categorical
can be numbered but numbers have no significance
nominal data
consists of numerical ranked data that is ranked according to some criterion; each rank is different from the others, but the differences might not be equal. Ex: pain ratings, (on a scale of 1-10)
ordinal data
consists of ranked data with intervals between each other being equal but with no meaningful zero point ex: IQ scores
interval data
like interval data but zero point is meaningful; highest level of measurement. Ex: most lab values, no negative values, height, weight, age
ratio data
acquisition of new knowledge for its own sake, without reference to potential practical use
basic research
directed toward solving immediate practical problems with functional applications and testing theories that direct practice
applied research
structured process of investigating facts and theories in medicine exploring connections, improving patient care
clinical research
