Research methods chapter 13
the position of the median
N+1/2 == what?
limitations of grouped data
(1) the mean may not be representative of the scores of the individual participants in a group, (2) plotting means may not describe the true nature of what is being studied-----example of
real upper limit
- the largest value that would be classified as falling into an interval.
unimodal, bymodal
1 hump is ___ 2 humps is ___
positive - Negative
A line that starts low and increases represents a ___ relationship between the variables. A line that starts high and decreases represents a ___ relationship.
grouped data
A major advantage is convenience. You have one score that characterizes an entire distribution (mean).
outlier
A value that is widely separated from the rest of the data is called an
measures of spread
Also called measures of dispersion (or variability).
statistical inference
Assess the reliability of your findings
frequency distribution
Consists of a mutually exclusive categories (classes) into which you sort the actual observed values.
test stat
Each statistical test has its own specific __ ___
measures of central tendency
Gives you a single score that represents the general magnitude of a distribution.
pie chart
If your data are proportions or percentages a ___ ___ can be used.
scatterplot
In observational (correlational) research where both variables are on the ordinal, interval, or ratio scales of measurement, ____should be used.
pie chart --- mutually exclusive
Place the name of each category and the percentage of individuals who are in that category on the chart. Categories should be __ ___
descriptive statistics
Provide simple summaries about the sample and the measures. Allows you to summarize an entire distribution of scores with a few numbers.
the bars touch.
Resembles a bar graph except
standard deviation
The most commonly used measure of dispersion
bar graph
There is a horizontal x-axis (abscissa) and a vertical y-axis (ordinate).
statistical inference
Uses descriptive statistics in an attempt to determine population parameters.
statistics
Variables are used to compute
midpoints
___of each class are plotted in a histogram
bar graph
a graph of the number of depressive symptoms (DV) reported by different racial groups (IV).
When the distribution not normal, the median may be a more appropriate measure of central tendency because it is less affected by skewness. Unaffected by extreme scores (outliers). Interpretation of the median does not require any assumptions about the interval properties of the scale (no assumption about the distance between numbers).
advantages of the median
It is a score that actually occurred in the data. Represents the largest number of people having the same score. Applicable to nominal data The probability that an observation drawn at random will be equal to the mode is greater than the probability that it will be equal to any other specific score Unaffected by extreme scores (outliers).
advantages of the mode
statistics
are quantitative measurements of a sample. ___ allow us to objectively evaluate data.
outliers
are sometimes caused by errors in recording the data but they are often real data points.
mean
average of the scores
grouped and individual
data are commonly used together.
Value may not exist in the data. There in not an equation to calculate the actual median (only its position) It is not stable from sample to sample like the mean.
disadvantages of the median
When data are placed in categories, the mode may change if data are grouped differently. It is not stable from sample to sample like the sample mean. May not be particularly representative of the entire collection of numbers, especially if the mode is 0. The actual mode can not be calculated with a formula.
disadvantages of the mode
symmetric
if you drew a line vertical line down the center of the histogram, the graph would look the same on both sides.
frequency distribution -- width
in a ___ ___ Categories or classes are the same ___
variable
is a property of a person, object, or event that can take on different values. They fall under 1 of 4 scales of measurement.
distrubution
is a set of scores on a particular variable of interest.
SD or standard deviation
is expressed in the same unit of measurement as the original scores. It is the square root of the variance.
variance or s squared
is the averaged squared deviation from the mean
interquartile range
is the difference between the 75th and 25th percentile
range
is the simplest and least informative (crudest measure of dispersion).
bar graph
is used to graph a nominal or categorical independent variable and a dependent variable that is on the ordinal, interval, or ratio scales of measurement.
histogram
is used to plot a frequency distribution.
negatively skewed
looks like the histogram has a tail to the left.
positively skewed
looks like the histogram has a tail to the right.
individual data
makes most sense when examining change over time.
statistic
may be thought of as a numerical index of some characteristic of the data.
mutually exclusive
means every data point will fall into only one category
measures of central tendency
mode , median & mean are all..
It can be calculated and manipulated with a formula If we drew many samples from one population, the sample means would be more stable (less variable) estimates of central tendency of that population than the sample medians or sample modes
name 2 advantages of the mean
Affected by extreme scores Value may not exist in the data Interpretation of means requires at least some faith in the interval properties of the data.
name 3 disadvantages of the mean
Symmetric Unimodal Bell-shaped Mean, median, and mode have the same value Has limits of negative infinity and positive infinity
name 5 characteristics of a normal distribution
measures of spread
range, variance, & SD are all...
median
the middle score when you put the scores in order
mode
the most frequent score in a distribution. It is the crudest measure of central tendency
real lower limit
the smallest value that would be classified as falling into an interval.
median
when the distrubution isn't normal the __ is more appropriate than the mean