Searching literature and statistics Flashcards
Strategies for reading research articles
- Read abstract
- Read discussion
- Read methodology
- Take notes, read 2-3 times, look up definitions as you read
Statistics definition
The methodology for collecting, analyzing, interpreting, and drawing conclusions from data
Data
facts that can be analyzed or used in an effort to gain knowledge or make decisions; info
Statistics provides methods for:
Design- planning and carrying out research
Description- summarizing/ exploring data
Inference- make prediction and generalize about phenomena represented by the data
Why should we care about statistics?
To be able to effectively conduct research- gives us a tool to use such that researchers can consistently analyze data- unbiased and accurate- honest, reliable
To be able to read and evaluate journal articles- statistics often show us both the questions and the answers to those questions
To further develop critical thinking and analytic skills
To act as an informed consumer
Variable
A characteristic, #, or quantity that incr or decr. Over time, or
takes different values in different situations; a factor in a scientific experiment that may be subject to change
Quantitative
numerical- can be counted/ measured
Qualitative
you can’t measure it- the quality- color, categories, etc
Descriptive stats
For organizing and studying data- graphs, charts, tables, averages, measures of variation, percentiles
Ex: Arithmetic mean: sum of a collection of #s divided by # of #s
Standard deviation: measure of spread of #s in a set of data from its mean value
Derived from sample variance
Why must we consider the standard deviation when comparing group means?
Two groups of data with the same mean may have very different standard deviations (spread of values- used to determine how much the mean accurately represents the data set)
Significance test
a method of inference used to support/ reject claims based on sample data
Null hypothesis
(H sub 0): no observed difference between experimental groups
Alternative hypothesis
(H sub A): there will be an observed difference between experimental groups
Statistical significance
a result is unlikely to have occurred by chance alone, and is determined by a p-value
P-value
an estimate of the probability that a particular result could have occurred by chance, assuming the null hypothesis is true
Analyzing the p-value
Ex: the probability of being wrong if you reject the null hypothesis, or the probability of finding a difference when there is no real difference
If the p-value is small, there is a low probability that you will be wrong if you reject the null hypothesis and accept the alternative
Estimate of the probability that a result was determined by chance- if the p-value is small, it is more likely that the results were not due to chance, and therefore actually significant to the experiment
Predictor variables
independent- variable that is being studied or manipulated in order to measure the result in the response variable
Response variables
dependent
Linear regression
statistical analysis assessing the association between 2 continuous variables
Example of linear regression:
can we predict body weight based on a person’s height
Null: slope of the data = 0
Alternative: slope doesn’t = 0- has either a positive or negative relationship
T-test
used to determine if 2 variables are significantly different from one another
Example of t-test
does insecticide treatment (categorical predictor variable) reduce insect damage (continuous response variable) to plants?
Null: no diff in the mean values between 2 samples
Alt: means are different- the treatment is reducing insect damage
Chi-square
used to determine whether there is a significant association between 2 categorical variables
Example of chi-square
Ex: are white and gray moths eaten equally by birds? The predictor variable is moth color (white or gray; categorical) and the response variable is mortality (dead or alive; categorical)
Null- variable A and B are independent
Alternative- variable A and B are NOT independent
