the 3 types of data are...
- nominal/categorical
- ordinal (ranked/relative data)
- interval (on a scale)
central tendency
a value that is representative of a data set
- 3 ways to measure: mean, median & mode
- shows how similar the data points are in a set of data
mean
an average of data points
median
the number in the middle
mode
the most frequently occurring measurement
range
the measure of the spread of data
(a.k.a. largest # - smallest #)
error bars
is a line that extends above and below a bar in a chart/data point in a graph
- shows range of data, standard deviation, 95% confidence intervals (CI), & standard error (how close to the means the samples are)
standard deviation (SD)
is used to summarize the spread of values around the mean
- normal distribution = 68% of all values lie within the 1st standard deviation
- compares the means & spreads the data between 2 or more samples
- can indicate that control variables are not const
what does the size of the standard deviation (SD) indicate about the data?
the SD shows the spread of the range
- small SD signifies a small range
- large SD signifies a big range
statistical significance
refers to how probable it is that a relationship is caused by pure chance
t-table
is based on probability (p) that chance alone could make a difference
t-test
is used to compare the means of 2 samples to determine whether they are statistically different
-
2 conditions
: data shows normal distribution (follows the bell-curve) & the sample size must be at least 10
-
3 things needed for the t-test
: df (degrees o
null hypothesis
a hypothesis that says that the IV & DV have no correlation; it's based on chance
critical value
-
if the critical value is ? 0.05
(5%) (in the p-value column), then the
data is not reliable
; you must
accept the null hypothesis
; the
difference is due to chance
-
if the critical value is > 0.05
, then the
data is reliable
; you must
reject the null
correlation v. causation
correlation
shows a linear relationship between 2 variables. however, just because there's a pattern doesn't mean one is the cause of the other. if the IV affects the DV, then it's
causation
.
accuracy
The degree of closeness of a measured value to its true amount.
aim
A short statement describing the purpose or reason for an experiment.
chi-squared test
A statistical test for determining the significance of departures of observed data from an expected results.
control
A standard (reference) treatment that helps to ensure that the responses to the other treatments can be reliably interpreted.
controlled variable
Variable that is fixed at a specific amount across all treatment groups.
correlation
A relationship in which variables vary together in some predictable way, but cause and effect in not implied.
Data
Facts collected for analysis.
dependent variable
a variable whose values are determined by another variable
descriptive statistics
Calculated values that summarize the main features of a collection of data.
error bars
a graphical representation of the variability of data. Used on graphs to indicate the uncertainty in a reported measurement.
graph
A diagram which often displays numerical information in a way that can be used to identify trends in the data.
hypothesis
A tentative explanation of an observation, capable of being tested by experimentation.
independent variable
A variable whose values are set, or systematically altered, by the investigator.
Median
the central value in a sorted data set
mode
the value that occurs most often in a data set
precision
the repeatability of a measurement
qualitative data
data described in descriptors or terms rather than by numbers
quantitative data
Data able to be expressed in numbers. Numerical values derived from counts or measurements.
regression
a relationship between a dependent variable and one or more independent variables
sample mean
Estimate of the true population mean based upon data collected by random sampling. Valid for population data that are normally distributed. The sum of the data divided by the number of data entries (n).
sample standard deviation
A calculated statistic expressing the variability of a sample population around its mean.
scientific method
The use of an ordered, repeatable method to investigate, manipulate, gather, and record data.
Statistic
A calculated measure of some attribute of a sample (e.g. the arithmetic mean)
Student's t-test
A test used to determine if the difference between the two sample means is significant.
table
A representation of data, often summarized, organized in rows and columns.
trend (of data)
a relationship between variables in a data set.
variable
A factor in an experiment that is subject to change.
What is the standard deviation?
how far a number deviates from the mean
The standard deviation is used to what?
summarize the spread of data around the mean
How is the standard deviation represented?
mean +/- SD
What does a large standard deviation mean?
data is widespread around the mean
What does a small standard deviation mean?
data is clustered around the mean
Name two applications of standard deviation?
1. used to compare the means of 2 populations
2. useful tool in evaluating experimental designs
What do large SDs across treatments show?
that variables may not be controlled
Most biological variation shows a normal distribution which displays what?
a bell-shaped curve
What percentage of data falls between +/- 1 SD of the mean?
68%
What percentage of data falls between +/- 2 SD of the mean?
95%
Error bars are used to show what?
the variability of data graphically
What are error bars used to represent?
standard deviation and the range
Correlation does not establish what?
causation
What do error bars show?
The range of data (all the samples), the standard deviation (68% of the sample), and the 95% confidence intervals