Professional Documents
Culture Documents
Psychological Statistics
Psychological Statistics
Psychological Statistics
Measures of Variability points; in contrast, the smaller the IQR, the more
bunched up the data points are around the mean.
Interquartile Range (IQR)
The IQR range is one of many measurements
Formula:
used to measure how spread out the data points in a
data set are. It is best used with other measurements
IQR = Q3 – Q1 such as the median and total range to build a
complete picture of a data set’s tendency to cluster
Q3 is the upper quartile and Q1 is the lower quartile
around its mean.
Odd Number Data
Calculating Outliers
1. Put the number in order.
1. Upper Fence = Q3 + (1.5 * IQR)
1, 2, 5, 6, 7, 9, 12, 15, 18, 19, 27
2. Lower Fence = Q1 – (1.5 * IQR)
2. Find the median
1, 2, 5, 6, 7, 9, 12, 15, 18, 19, 27 Use your fences to highlight any outliers, all values
3. Place parentheses around the numbers above that fall outside your fences.
and below the median. Not necessary
statistically, but it makes Q1 and Q3 easier to - Outliers are extreme values that differ from most
spot. values in the dataset. You find outliers at the
(1, 2, 5, 6, 7), 9, (12, 15, 18, 19, 27) extreme ends of your dataset
4. Find the Q1 and Q3 - Outliers can have a big impact on your statistical
Think of Q1 as a median in the lower half of analyses and skew the results of any hypothesis
the data and think of Q3 as a median for the test if they are inaccurate.
upper half of data. - These extreme values can impact your statistical
(1, 2, 5, 6, 7), 9, ( 12, 15, 18, 19, 27) power as well, making it hard to detect a true
Q1=5, Q3=18 effect if there is one.
5. Subtract Q1 from Q3 to find the interquartile - It’s best to remove outliers only when you have a
range. sound reason for doing so.
IQR = 18 – 5 = 13 - Some outliers represent natural variations in the
population, and they should be left as is in your
Even Number Data dataset. These are called true outliers.
- Other outliers are problematic and should be
1. Put the numbers in order.
removed because they represent measurement
3, 5, 7, 8, 9, 11, 15, 16, 20, 21
errors, data entry or processing errors, or poor
2. Make a mark in the center of the data
sampling.
3, 5, 7, 8, 9, | 11, 15, 16, 20, 21.
3. Place parentheses around the numbers above Shape pf a Frequency Distribution
and below the mark you made in Step 2 – it
makes Q1 and Q3 easier to spot. 1. Symmetrical Distribution
(3, 5, 7, 8, 9), | (11, 15, 16, 20, 21) o in a symmetrical distribution, it is
4. Find Q1 and Q3 possible to draw a vertical line
Q1 is the median (the middle) of the lower through the middle so that one side of
half of the data, and Q3 is the median the distribution is a mirror image of
(middle) of the upper half of the data. the other.
(3, 5, 7, 8, 9), | (11, 15, 16, 20, 21).
Q1 = 7 and Q3 = 16.
5. Subtract Q1 from Q3 to find the interquartile
range.
IQR = 16 – 7 = 9
Percentile
FORMULA:
Hypothesis Meta-analysis
- Prediction, often based on informal observation, - is an important development in statistics that has
pervious research, or theory, that is tested in a had a profound effect on psychology, especially
research study. clinical and social psychology, and on many other
o Alternative Hypothesis - it is a claim scientific fields, such as medicine, public health,
about the population that is education, and organizational behavior. This
contradictory to null hypothesis and procedure combines results from different
what we conclude when we reject null studies, even results using different methods of
hypothesis. measurement. When combining results, the
o Null Hypothesis - statement about a crucial results combined are effect sizes.
relation between populations that is 2 Types of Correlation
the opposite of the research
hypothesis; statement that in the 1. Positive (+) Correlation
population there is no difference (or a o Directly proportional
difference opposite to that predicted) o If one variable increases, the other
between populations; contrived increases too. And if one variable
statement set up to examine whether decreases, the other decreases as
it can be rejected as part of well.
hypothesis testing. (is often used) 2. Negative (-) Correlation
o Inversely proportional
Hypothesis Testing
o If one variable increases, the other
- Procedure for deciding whether the outcome of a decreases. And if one variable
study (results for a sample) supports a particular decreases, the other increases
theory or practical innovation (which is thought to
Purpose of Correlation
apply to a population).
- Used to measure and describe the relationship
Errors in Hypothesis Testing
between two variables
o Type I - To figure out which variable are connected or
False positive related.
Rejects a null that is true
Example:
o Type II
False negative
Accepts a null that is false
Example:
Effect Size