Download as pdf or txt
Download as pdf or txt
You are on page 1of 36

Psychology 253

Data Analysis in Psychology


Why?
~ Understand psychology phenomena
~ Prevalence
~ Relationships between variables
~ Test efficacy of interventions

Chapter 1: Introduction to Statistics


Scales of Measurement
~ Any data collection requires that we make measurements of our observations.
~ It involves...
o The categories used to measure a variable, makes up a scale of measurement.
o The relationships between these categories determines the different types of scales.
~ Measurement assigns individuals or events to categories
o The categories can be names, such as male/female or employed/unemployed
o They can be numerical values, such as 68 inches or 175 pounds
~ The complete set of categories makes up a scale of measurement
o Relationships between the categories determine different types of scales
There are 4 different types of scales…
Scale Characteristics Examples
Nominal •Label and categorize •Gender
•No quantitative distinctions •Diagnosis
•Experimental or Control
Ordinal •Categorizes observations •Rank in class
•Categories organized by size or •Clothing sizes (S,M,L,XL)
magnitude •Olympic medals
Interval •Ordered categories •Temperature
•Interval between categories •IQ
of equal size •Golf scores (above/below
•Arbitrary or absent zero point par)

Ratio •Ordered categories •Number of correct answers


•Equal interval between •Time to complete task
categories •Gain in height since last
•Absolute zero point year

~ This material is arguably in the “Top Ten Most Important” concepts the students will encounter in the
study of statistics and may merit identifying it as such.
~ The word nominal means having to do with names or labels.
o It involves classifying individuals or events into categories.
o They all have different names but are not related in any way.
o For example: if you were measuring academic majors for a group of university students, those
majors would be classified according to categories such as: psychology, sociology, business
science, biology etc. so each student in that group would be classified in a category.
~ An ordinal scale consists of categories which are organised in a sequence.
o So, measurements are ranked in class.
o Sizes etc.
o For example: in a psychology module, one could rank students according to who came first,
second, third or fourth in the course in terms of academic achievement. The fact that
categories form an ordered sequence means that there is a relationship between categories.
~ An interval scale consists of ordered categories that are all intervals of exactly the same size.
o So, it is characterized by equal intervals between scale units.
o Basically, the difference between two values are meaningful.
o For example: Person A scored 65% in her psychology 253 exam and person B scored 80% in
her psychology 253 exam. We can also then say that person B scored 15% higher than person
A and we can go further and say that it is equal to Person C scoring 70% and Person D scoring
85% (the difference between both sets of scores is equal). When thinking of this in terms of a
scale, the zero point of the scale is arbitrary, in other words it has no zero point.
~ The ratio scale is basically an interval scale with and added characteristic.
o It has an absolute zero which means that a score of zero indicates none of the variable being
measured.
o For example: number of correct answers on a student’s psychology 253 exam. It can either be
any number of correct answers as well as zero correct answers.
Three Data Structures, Research Methods, and Statistics
Data Structure I:
o Descriptive research (individual variables)
o One (or more) variables measured per individual
o “Statistics” describe the observed variable
o May use category and/or numerical variables
~ Data structure 1: One or More Separate Variables Measured for Each Individual:
~ Descriptive Research:
~ This involves measuring one or more separate variables for each person or participant.
~ The intention here is to simply describe as the title says.
~ Think of this example
Individual Number of hours Number of hours Number of hour
exercise in a day sleeping in a day studying in a day

A 2 6 4

B 1 7 3

C 3 4 5

D 4 8 2
~ So in this table we can see that it speaks to how many hours an individual exercises, sleeps and
studies in each day.
~ Here it is different for each person, and we are simply describing the variables by saying that
Person A exercises for 2 hours a day, sleeps for 6 hours and studies for 4 hours each day, and so
on with the rest of the individuals described here.
Relationships between variables
~ Is very important.
o Two (or more) variables observed and measured
o One of two possible data structures used to determine what type of relationship exists
~ Most research aims to examine whether there is a relationship between variables. For example:
~ Is there a relationship between number of hours spent studying a day and the results on a test?
~ Is there a relationship between number of hours spent studying for psychology 253 exam and the
results of the exam?
~ To establish whether there is a relationship – we first have to make observations of the variables.
Data Structure II:
o The correlational method
o One group of participants
o Measurement of two variables for each participant
o Goal is to describe type and magnitude of the relationship
o Patterns in the data reveal relationships
o Non-experimental method of study
~ Data Structure 2 involves the correlational method.
~ So, this is observing one group with two variables being measured.
~ So we just spoke about examining the relationship between variables and this can be done using two
variables.
~ Simply put, we can measure two variables for each individual.
~ You have examples in your textbook but I want to focus on another example here:
~ In the correlational method two different variables are observed to determine whether there is a
relationship between them.
~ Example
STUDENT NUMBER OF HOURS ACADEMIC
SPENT ON SOCIAL PERFORMANCE
MEDIA (Results on a test)
A 5 50%

B 6 40%

C 4 60%

D 2 70%

o This table shows information on 4 students in relation to the amount of time they spend on
social media and the results they score on a test.
o So, you can see that person C spends 4 hours on social media and scored 60% on her test
o Whereas person B spends 6 hours on social media and scored 40% on his test…
o draw scatterplot with whiteboard function.
Figure 1.5 Data structures for studies evaluating the relationship between variables
o One of two data structures for studies evaluating the relationship between variables.
o Note that there are two separate measurements for each individual (Facebook time and academic
performance).
o The same scores are shown in table (a) and graph (b).

Correlational Method Limitations


o Can demonstrate the existence of a relationship
o Does not provide an explanation for the relationship
o Most importantly, does not demonstrate a cause-and-effect relationship between the two
variables
o So, as with any technique there may be some limitations.
o Regarding the correlational method that we just heard, we can deduce that the results of a
correlational study can demonstrate the existence of a relationship between two variables, but
they do not explain the relationship.
o More noteworthy is that a correlational study cannot demonstrate and cause and effect
relationship between the two variables.
o So, as I mentioned in the example before the table shows that the more hours spent on social
media resulted in a lower percentage scored on a test, however there could also be other factors
involved that influenced that student or individual to score low on his or her test that day.
o Hence, we can’t conclude to say that if you spend less time on social media your academic
performance would increase…
Data Structure III:
o Comparing two (or more) groups of scores
o One variable defines the groups
o Scores are measured on second variable
o Both experimental and non-experimental studies use this structure
~ Instructors may wish to introduce the term “quasi-experimental” in this section.
~ The third data structure is: Comparing two or more groups of scores: Experimental and Non-
experimental methods:
~ In this data structure, you use one variable to define groups and then measuring the second variable
to obtain scores for each group.
o Example: You have two groups… As mentioned in the text book: 1 group is exposed to a
violent video game and another group exposed to a non-violent video game. The results can
be seen in the table on the next slide…
Figure 1.6: Data structure for studies comparing groups]

o The second data structure for studies evaluating the relationship between variables.
o Note that one variable is used to define the groups and the second variable is measured to
obtain scores within each group.
o So here you are assessing two variables, type of video game (violent and non-violent) and the
second variable - aggressive behavior is measured to obtain the scores as seen in this table
above.
Experimental Method
~ The experimental method is systematic and scientific approach in which the research manipulates one
or more variables.
~ So, one variable is manipulated and the other is controlled.
o For example: administering a diet supplement to one group and the other group receives a
placebo; there weight-loss is being measured.
~ Goal of experimental method
o To demonstrate a cause-and-effect relationship
o This requires manipulation by the researcher changing its value from one level to another.
~ Manipulation
o The level of one variable is determined by the experimenter
o So relating back to the previous example about violent video games and non-violent video
games.
o The research then manipulates by giving one group of boys a violent video game to play and
the other group non-violent video game.
o The second variable as mentioned is the variable that is measured and in this example it is
whether the manipulation had an effect or not.
~ Control rules out influence of other variables
o Participant variables
o Environmental variables
o Regarding control, this is where the researcher makes sure that no external factors influence
the relationship.
o So in the example of the diet supplement where one group receives the diet pill and the other
group receives the ‘fake’ diet pill (placebo), the researcher then makes sure that no extraneous
factors influence the relationship by ensuring maybe the both groups do the same exercise
routine and eat the same foods etc.
Experimental Method: Control
~ Within the experimental method, there can be controlled and experimental conditions.
~ Individuals in a controlled condition do not receive the experimental treatment (for example, the
group of boys who played the on-violent video games or the one group of people who received the
placebo diet pill.
~ Methods of control
o Random assignment of subjects
o Matching of subjects
o Holding level of some potentially influential variables constant
~ Control condition
o Individuals do not receive the experimental treatment
o They either receive no treatment or they receive a neutral, placebo treatment
o Purpose: to provide a baseline for comparison with the experimental condition
o The purpose of such a condition is to provide a baseline for comparison.
~ Experimental condition
o Individuals do receive the experimental treatment
o Individuals in the experimental conditions do receive the experimental treatment (i.e the
violent video games or the diet pill)
Independent/Dependent Variables
~ Independent variable is the variable manipulated by the researcher
o Independent because no other variable in the study influences its value
~ Dependent variable is the one observed to assess the effect of treatment
o Dependent because its value is thought to depend on the value of the independent variable
~ So in our example the independent variable would be the amount of violence (so the violent vs non-
violent video games) and the dependent variable is the level of aggressive behaviour.
Non-Experimental Methods
~ Non-equivalent groups
o Researcher compares groups
o Researcher cannot control who goes into which group
~ Pre-test / Post-test
o Individuals measured at two points in time
o Researcher cannot control influence of the passage of time
~ Independent variable is quasi-independent
~ There are also non-experimental methods.
~ The first being no-equivalent groups:
o Any study that follows a scientific requirement such as the previous examples are known as
experimental studies.
o And as such there are also designs that are no-experimental but still examine the relationships
between variables.
~ A pre-test and post-test study uses time such a s before and after to create groups of scores. Lets
see the examples…
Two examples of non-experimental studies: Figure 1.7
~ Two examples of non-experimental studies that involve comparing two groups of scores.
~ In (a), a participant variable (gender) is used to create groups, and then the dependent variable
(verbal score) is measured in each group.
~ In (b), time is the variable used to define the two groups, and the dependent variable (depression)
is measured at each of the two times.
Chapter 2: Frequency Distributions
Frequency Distributions
~ A frequency distribution is
o An organized tabulation
o Showing the number of individuals located in each category on the scale of measurement
~ Terminology associated with frequency distributions is one of the least “standardized” across
disciplines and texts students might encounter.
~ Instructors may wish to emphasize the importance of being precise with the terms provided by the
text authors, but also be aware that terms may differ in other texts or courses.
~ Can be either a table or a graph
~ Always shows:
o The categories that make up the scale
o The frequency, or number of individuals, in each category
~ I am not going to go into any definitions as you as the student needs to make sure you go through
every concept and make sure you become familiar with the terminology used in statistics.
~ This is crucial…
o So we know that a frequency distribution is basically data organized in a table that can explain
something…
o It can also be distributed onto a graph like we saw in the previous podcast on data structures.
o But here you will learn the statistical input of data in tables or graphs. Like I said before
practice makes perfect, so go through your textbook and practice using the Learning Check
questions.
o So you can know more or less how you may be examined or what a certain questions asks of
you.
o In any frequency distribution it indicates the categories which make up the scale as well as the
frequency or number of individuals in each category.
Example
X f
8 9 8 7 10 9 6 4 9 8
7 8 10 9 8 6 9 7 8 8 10 2

9 5

N= (number of scores 8 7

20) 7 3
X= score
6 2
f= number of times
5 0
scores occurred.
4 1

~ Here you have a set of scores right… there are 20 scores.


~ The sum of the scores are denoted with the N symbol.
~ We say here N equals 20 as there are 20 scores.
o In statistics, a frequency distribution takes a set of scores (and usually we organise the scores
from highest to lowest) but it takes that set of scores and places it in a table.

ΣX = N
X f

5+4+4+3+3+3+2+ 5 1

2+2+1= 29 4 2

3 3

ΣX = 29 2 3

1 1

ΣX2= 97
o see this table here… we have 20 scores but we can see that some scores present more than
once so we can group it as follows…
o we can see that N= sum, X=score and f=frequency (number of times score occurred). Again
here you must basically memorize these symbols because they are used throughout stats.
§ So these set of scores are organised in the table under the column X (excluding the
scores that occur more than once) and remember it must always be tabulated in order
from highest to lowest.
§ Once you organise your scores from highest to lowest you look at each number and
see how many times a score of 10 occurred, a score of 9 occurred and so on…
§ Here you actually count how many timesa score of 10 occurred etc..
§ You can now see clearly the scores in the table. 2 people scored the highest score
which is 10 and you can also see that no one scored 5 but it is included in the table.
§ Remember the 4 different scales that we spoke about previously, with an ordinal,
interval and ratio scale the categories are listed in order rom highest to lowest.
Frequency Distribution Tables
~ Structure of frequency distribution table
o Categories in a column (often ordered from highest to lowest but could be reversed)
o Frequency count next to category
~ Σf = N
~ To compute ΣX from a table
o Convert table back to original scores or
o Compute ΣfX
~ As mentioned, the categories are ordered from highest to lowest and they are displayed under the
column X and the number of times that score appeared is found in the column f.
~ But now we want to know the total numbers of scores in the distribution and this is denoted by the
symbol above.
o to calculate the sum of scores in the distribution you need to look at both columns.
o Let’s look at the next example…
Calculating the scores in a Frequency Distribution
Consider this table…
X f ΣX

5 1 5

4 2 8

3 3 9

2 3 6

1 1 1

ΣX = 29

~ You see a set of scores and how many times the score appeared which is the frequency.
~ So to calculate the sum of the scores we do the following:
o Remember I said in the previous slide you need to consider both columns when calculating
the sum…
~ You will write out each score including the amount of times each score appeared:
~ 5+4+4+3+3+3+2+2+2+1= 29
~ To get the sum on the frequency squared you square each score and add the squared values = 97
This you do on a calculator…
~ Another way to get the sum of X and input in in a table is as follows…
~ Let’s input into the table together…
Proportions and Percentages
Proportions
• Measures the fraction of the total group that is associated with each score
f
proportion = p =
N
• Called relative frequencies because they describe the frequency ( f ) in relation to the total
number (N)
Percentages
• Expresses relative frequency out of 100
f
percentage = p(100) = (100)
N
• Can be included as a separate column in a frequency distribution table
~ The ability to quickly and comfortably convert between fractions (proportions), decimal fractions
(relative frequency), and percentages is fundamental to success in this course.
~ Some students struggle with reconciling the fact that although these are three distinct metrics, they
all point to the same “deep” meaning.
~ There are other measures that describe the distribution of scores which we can also interpret.
~ These are Proportions and Percentages.
§ Proportion measures the fraction of the total group associated with each score so for
example we see in the previous table that two people scored 4 so we can say that 2
out of 10 people had 4, this is then demonstrated in this form above.
~ Researchers also use percentages to describe a distribution, and this can be done by first finding the
proportion and then multiplying that by 100.
§ Use whiteboard to show proportion and percentage...
Example 2.4: Frequency, Proportion and Percent

Grouped Frequency Distribution Tables


~ If the number of categories is very large, they are combined (grouped) to make the table easier to
understand
~ However, information is lost when categories are grouped
o Individual scores cannot be retrieved
o The wider the grouping interval, the more information is lost
~ Today’s podcast will cover grouped frequency tables…
~ In the previous podcast we had a distribution of 20 scores etc, but what about the distributions with
very large categories? This is when grouped frequencies come in…
o when we have a very large range of values, we cannot list all scores in a table we then turn to
grouped frequencies.
o Our aim is to always gain a simple organised picture of a distribution, that is why we use
tables.
o so a grouped frequency can be done by grouping scores into intervals and then listing those
intervals in a table instead of each individual score.
§ So for examples we can group the number of scores of students who had marks in the
80s and 90s etc… this table is called a group frequency distribution table.
§ We can see an example in the next slide.
Example
~ In this table you can see that instead of listing individual scores.
~ When you have large distributions you can group them as follows.
~ Deciding on how to group these scores involves following some guidelines.
~ So you will have the full set of score in front of you and then you create the table with the intervals
based on the scores range.
~ Then you will look at the scores and see how many scores from the list fit in to each interval.
~ So for example, 3 of the scores ranged between 90 to 94. (The actual scores were 91; 94; 93).
X f
90-94 3
85-89 4
80-84 5
75-79 4
70-74 3
65-69 1
60-64 3
55-59 1
50-54 1

“Rules” for Constructing Grouped Frequency Distributions


~ Requirements (Mandatory Guidelines)
o All intervals must be the same width
o Make the bottom (low) score in each interval a multiple of the interval width
~ “Rules of Thumb” (Suggested Guidelines)
o Ten or fewer class intervals is typical (but use good judgment for the specific situation)
o Choose a “simple” number for interval width (e.g., 2,5,10)
~ This slide shows a recommended treatment for the four guidelines presented in the text.
~ The text presents it slightly differently by indicating these are “guidelines” rather than absolute
requirements.
~ However, violating guideline 4 distorts the information conveyed and violating guideline 3 makes it
much more difficult to assimilate the information conveyed by the table.
~ Consequently, each instructor should clarify expectations for her class: are these “guidelines” or
“rules?”
~ So here we want to give you some guidelines in terms of grouping frequencies.
~ Firstly, the grouped frequency distribution table should have about 10 intervals (Go back to table and
show them).
~ Remember the whole purpose of a frequency distribution table to is help the researcher or reader to
see the data in an organized manner.
~ Second guideline is that the width or size of the intervals should have a simple number such as 4 or 5
or 10 or lower even like 2.
~ Here you need to think what would be easy to count in so 5s or 10s seem relatively easy to count.
~ Thirdly, the bottom score in each class interval should be a multiple of the width.
o For example, if you are using a width of 10 points then the intervals should start with 10, 20,
30 40, etc.…
~ The fourth guideline is that all intervals should be the same width.
~ This is to ensure that any score will fit or belong into these intervals.
o refer to the table example: here you can see that the width of 5 was sufficient to group these
scores.
o remember the wider the intervals are the more information is lost so you want to keep it as
concise ad possible because in a group frequency table you can’t tell exactly what the score is.
The shape of a Frequency Distribution
~ We get an idea of the shape of the distribution
~ Can enhance the understanding of the nature of the data
~ There are many different ways you can input data from tables onto graphs, you would have come
across the various types of graphs such as bar graphs, scatterplots etc.
~ But what is almost more important is what we can see and explain from looking at a graph.
~ Here, the shape of the frequency distribution on the graph is very important.
~ There are three characteristics that completely describe any distribution and these are:
1. Shape: we classify the graphs according to the shape, so whether they are symmetrical or
skewed.
2. Central Tendency: this measure where the centre of the distribution is located
3. Variability: this is relating to the degree of whether the scores are spread over a wide range or if
they are clustered at one point.
Figure 2.9 – IQ Population Distribution Shown as a Normal Curve

~ The population distribution of IQ scores: an example of a normal distribution.


~ In this graph you can see something that is called a normal curve.
~ A normal curve is where the greatest frequency is in the middle and smaller frequencies move to
what is called the extreme ends.
~ In this example of IQ scores distributed on this graph, it then tells us that majority of people’s IQ
scores that were tested are found in the middle.
~ The extreme scores then lie on either ends.
~ It is extreme for the fact that it shows few people had a very low IQ score of less than 70 and on the
other extreme we can see that few people had IQ scores of higher than 130.
Figure 2.10 Distribution Shapes
~ Examples of different shapes for distributions.
~ As I mentioned in the first slide, we identify shape according to symmetry and skewness.
~ In a symmetrical distribution it is possible to draw a straight line through the middle of the graph so
that one side is identical to the other… (show them the line…)
~ Then, with skewness, this is also important in that it tells us where scores tend to pile up at one point
of the graph and taper off gradually towards the other end.
~ This is what we call either positively or negatively skewed.
~ Any time you see a skewed distribution to the right side of the graph then it is a positively skewed
distribution because the tail of the graph points to the positive on the x-axis.
~ Any time you see a skewed distribution the left-hand side of the graph then it is termed a negatively
skewed distribution.
~ For example, if you have a group of students who wrote a difficult exam, you will know that majority
of the scores would be low with only a few individuals scoring high marks this would produce a
positively skewed distribution.
~ In a similar instance where students write and easy exam, majority of the students will score high and
few score low and this in turn shows a negatively skewed distribution.
Example: Negatively Skewed

Example: Positively Skewed


Chapter 3: Central Tendency
Figure 3.1: Locate Each Distribution “Center”
~ Three distributions demonstrating the difficulty of defining central tendency. In each case, try to
locate the “center” of the distribution.
~ This slide presents a good opportunity for a discussion of how central tendency should be
characterized and how the same measure of central tendency may not be suitable for different
distributions
~ We know that central tendency is a statistical measure to determine which single score defines the
center of any distribution which then is the most representative of the entire group of scores.
~ This is easily seen on a normal distribution, but it becomes tricky when we have positively and
negatively skewed distributions.
~ In the las podcast we spoke about the different shapes of distributions but now we are also able to
calculate the mean of distributions.
~ You can see on the graphs here that there are different distributions.
~ A normal, a negatively skewed distribution and the bottom one which is normal but has two sets of
distributions.
~ To accurately measure central tendency we can calculate the mean, median and mode.

Central Tendency Measures


~ Figure 3.1 shows that no single concept of central tendency is always the “best”
~ Different distribution shapes require different conceptualizations of “center”
~ Choose the one which best represents the scores in a specific situation
~ As mentioned now, the goal of central tendency is to provide the best representation of scores in a
distribution. Now we can look at calculating the mean…
The Mean
~ The mean is the sum of all the scores divided by the number of scores in the data
~ Population:

µ=å
X
N
~ Sample:

M=
åX
n
~ Instructors my wish to have students compare and contrast these two formulas.
~ The mean is basically the average…
~ To calculate the mean we all all the scores and divide by the number of scores in the data.
~ The formula to calculate the mean is shown here.
~ There are two different ways to calculate the mean firstly in terms of the population and secondly for
a sample.
~ Here you can spot the slight differences between the two formulas.
~ For the formula to calculate the mean for the population it is as follows:
~ So this symbol represents the population mean and it is calculated by adding all scores and dividing
by the number of scores. So the sum of X over N.
~ For the sample mean the symbol is M=the sum of x over the number of scores…
~ So if you see a mean being represented by and M then you are dealing with a sample mean.
~ If it it being represented by the U symbol, then you are dealing with a population mean.
~ Lets looks at an example:
Example:
~ Calculating the population mean:
5, 6, 7, 3, 4
~ N= 5 scores
~ Add up all scores first: 5+6+7+3+4= 25
~ The mean is: 25
o So remember we first count how many scores there are, we see that there are 5 scores so N=5
o Then we add up all the scores which gives us 25
o Now to calculate the population mean we say pop mean = 25 divided by 5 =5 therefore our
population mean is 5.
~ Calculating the sample mean:
3, 6, 5, 3, 4
~ N= 5 scores
~ Add up all scores first: 3+6+5+3+4= 21
~ The mean is:
~ Gym attendance per week
The Weighted Mean
~ Combine two sets of scores
~ Three steps:
o Determine the combined sum of all the scores
o Determine the combined number of scores
o Divide the sum of scores by the total number
of scores
Overall Mean =

M=
åX +åX 1 2

n1 + n2
~ So what if we want to now know the mean of two sets of scores?
~ This is where we use the weighted mean formula.
~ So this formula is the Overall Mean equals the sum of scores for group 1 plus the sum of scores for
group 2 divided by the number of scores for group one and the number of scores for group two.
~ Now let’s see this in an example:
Example:
~ Hours of studying for a test per day:
~ Group 1: 3, 4, 5, 2, 6
~ Group 2: 2, 5, 3, 1, 5
~ So here we have a set of scores which indicate the number of hours a day two groups of 5 people
each, study for a test.
~ So we want to find out what the average or mean hours both groups studied for a test.
~ Group 1 hours are 3,4,5,2,6
~ Group 2 hours are 2,5,3,1,5
~ So let’s follow the steps we just spoke about for calculating the weighted mean:
~ Firstly lets add the total scores for each group: Group 1: 20 so sum of x =20
~ Group 2: 16 so the sum of x=16
~ And we know that there are a total of 5 scores per group so now lets write this out…
Computing the Mean from a Frequency Distribution Table
Quiz Score (X) f fX
10 1 10
9 2 18
8 4 32
7 0 0
6 1 6
Total n = Σf = 8 ΣfX = 66

M = ΣX / n = 66/8 = 8.25

~ We can also compute the mean from looking at a table.


~ So if a set of scores are inputted in a table, the calculation is usually easier to look at the column for X
and the column for F.
~ So we can see that there is one 10, two 9s, four 8s and one 6.
~ Our step is to add them all up so 10+9+9+8+8+8+8+6= 66
~ Next we can see that we have a total of 8 scores…
~ So our mean is inputted as follows: M= Sum of X divided by N which is 66 divide by 8= 8.25 therefore
the weighted mean is 8.25
The Median
~ The median is the midpoint of the scores in a distribution when they are listed in order from smallest
to largest
~ The median divides the scores into two groups of equal size
~ The second measure of central tendency is the median.
~ This has to do with finding the midpoint in a distribution.
~ So if you receive a list of scores it first needs to be organised from smallest to largest.
~ You look for the score that lies directly in the middle of the distribution of scores.
Example 3.7: Locating the Median (odd n)
~ Put scores in order
~ Identify the “middle” score to find median
3 5 8 10 11
“Middle” score is 8 so median = 8
~ So here for example you have a set of scores, you then order the scores from lowest to highest.
~ Now we look for the score that lies in the middle and has equal number of scores on both ends.
~ You can see that there are two scores on either side of this set of scores which then allows us to see
that the middle score is 8.
~ But what if we have a list of scores that is an even number?
Example 3.8: Locating the Median (even n)
~ Put scores in order
~ Average middle pair to find median
1 4 5 7 9
(4 + 5) / 2 = 4.5
~ Instructors may also want to point out the 4.5 is exactly ½ the distance between 4 and 5 on interval
or ratio measurement scales.
~ So you follow the same process in listing the scores from lowest to highest.
~ You can see we have 6 scores in total which is an even number.
~ We would then look at the two scores that lie in the middle.
~ Now we take those two score and add them up and then divide it by 2. this will give us the median.
So 4+5=9 and 9 divide by 2= 4.5
~ Therefore 4.5 is the median for this list of scores.
Finding the Precise Median for a Continuous Variable
~ A continuous variable can be infinitely divided
~ The precise median is located in the interval defined by the real limits of the value.
~ We must determine the fraction of the interval needed to divide the distribution exactly in half.
number needed to reach 50%
fraction =
number in the interval
~ Students have sufficient difficulty with this concept to justify developing another example similar to
the one the authors provided in Figure 3.4.
~ So what if we also want to find the precise median for a continuous variable?
~ This can be a little tricky.
~ We know a continuous variable consists of categories that can be split into fractional parts.
~ We must determine the fraction of the interval needed to divide the distribution in exactly half.
~ Such as time for example can be split into seconds, 10th of a second and hundredth of a second.
~ So by splitting it into fractional parts we can find the median by locating the precise point that
separates the bottom 50% of the distribution by the top.
~ So to calculate the fraction we use this formula:
~ Fraction= number needed to reach 50% divided by the number in the interval
~ Lets see an example…
Example
2, 3, 4, 4, 4, 5
~ Between first and second 4. three of the numbers are below and three are above right as that is the
definition if the median, so I want to show you that there is a number between 4 and 4.
~ So lets use an example here.
~ We ask 6 people how far it is that they have to walk from their house to the nearest shop in their
neighborhood…
~ So the first person said 2km, 3km, 4km etc.
~ Sometimes when you ask people how far they live they don’t always give you the precise number or
distance.
~ So lets put these numbers on a number line so that we can see what we are talking about
The Mode
~ The mode is the score or category that has the greatest frequency of any score in the frequency
distribution
o Can be used with any scale of measurement
o Corresponds to an actual score in the data
~ It is possible to have more than one mode
~ Some instructors may wish to extend the discussion of bimodal (and multimodal) distributions to
include the concept of major and minor modes to recognize that even when there is technically only
one “most frequently occurring” score, it might be helpful and useful to report more than one mode
to help better characterize the distribution.
~ The mode is basically the score that appears the most or has the greatest frequency…
~ The mode is used because it can determine the typical or most frequent value for any scale of
measurement, so for example we can say a black pen is the mode for students in PSY253 class as
most of the students write with a black pen.
~ This is determined from the amount of test or exam papers we receive.
Example
1,3,4,2,3,3,5,6,5,6,6,7,8,6,7,7,9,10,10,9,10
Mode: 6 as it appears four times
Shop f

Checkers 32
N=100 scores
Mode= 32
Woolworths 18

Pick ‘n Pay 28

Shoprite 22

~ So a basic example is this set of scores, you can see that 6 appears the most (4 times) hence 6 is the
mode.
~ Lets take another example by looking at the table above, you can see the different stores.
~ So for example a sample of 100 students on campus were asked which store they shop at for their
groceries, the most frequent answer is Checkers as 32 out of 100 students said they shopped at
Checkers.
~ It is also important to note that sometimes there can be two modes in a frequency.
~ Bimodial: is a distribution with 2 modes and a distribution with more than 2 modes is called
multimodial.
Chapter 4: Variability
Defining Variance and Standard Deviation
~ Most common and most important measure
of variability is the standard deviation
o A measure of the standard, or average, distance from the mean
o Describes whether the scores are clustered closely around the mean or are widely scattered
~ Calculation differs for population and samples
~ Variance is a necessary companion concept to standard deviation but not the same concept
o The twin concepts of variance and standard deviation are among the most challenging
concepts in a basic statistics course to communicate and to learn.
o Instructors will almost certainly want to invest special care in the preparation of materials to
help communicate these very difficult concepts.
o So in today’s podcast we wil focus on defining variance and standard deviation.
o The standard deviation is the standard or average distance ffrom the mean, its going to give us
a description of whether scores are close to or cluster together on the mean or widely
scattered out.
o Now we want to note that here too lik when calculating the mean there were slightly different
formulae, so here to calculate the SD of the population and samples are different too.
o Please ensure to familiarize yourself with the formulae in your text book as it also clearly
differentiates between sample and population formulae.
o Now Variance is a somewhat companion concept of SD. BUT it is not the same concept.
o So lets go through these formulae.
~ Step One: Determine the deviation
~ Deviation is distance from the mean
Deviation score = X − μ
~ Step Two: Find a “sum of deviations” to use as a basis of finding an “average deviation”
o Two problems
§ Deviations sum to 0 (because M is balance point)
§ If sum always 0, “Mean Deviation” will always be 0.
o Need a new strategy!
~ Having students try to come up with an intuitive method for developing a measure of variability
based on deviation scores is a great way to get them thinking about what a dead end strategy
averaging deviations is (because, of course, the average of deviations from the mean must be 0).
~ Several teams working on it in a classroom exercise often results in a valuable insight about the issue
(averaging absolute value of deviations) and might produce the one we use—squaring the deviations
to eliminate the negative values.
~ When we calculate the SD we basically asking how different the scores are from one another. So
deviation as mention means distance.
~ So for example: think of a psy 253 exam where the average grade is 70% but your score is 80%, so
essentially your score deviates from th mean by positive 10 points and likewise if you score below 80
you areminus 10 points from the mean.
~ So we can calculate for everyone in the class and calculate the deviation score for each person and at
the end we can calculate the average score for all deviations in the class.
~ This is essentially what SD is. It tells us the average distance of the sample from the pop mean.
Example
Finding the deviation score:
Deviation =x-µ µ=50 x=53
=53-50
Deviation score=3
Deviation =x-µ µ=50 x=45
=45-50
=-5
To get rid of the (-) you have to Ö (square) each and every score and then calculate the variance.
~ Step Two Revised: Remove negative deviations
o First square each deviation score
o Then sum the squared deviations (SS)
~ Step Three: Average the squared deviations
o Mean squared deviation is known as “variance”
o Variability is now measured in squared units
Population variance equals mean (average)
squared deviation (distance) of the scores
from the population mean
o The concept of sum of squared deviations (SS) is absolutely vital to efficient understanding of
the statistical tests presented in the remainder of the text.
o The authors have reduced the computational complexity and the cognitive load required of
students—contingent upon grasping and retaining the concept of SS presented in this chapter.
o The authors also lay the foundation for efficiently learning the fundamentals of ANOVA—
contingent upon grasping and retaining the concept of variance presented in this chapter.
o Consequently, this chapter is essential to success in the remainder of the course.
o If you have any negative scores you simply first square each deviation score and then add up
the squared deviations.
o So the next step would be to calculate the variance which is basically equal to the squared
deviations, soit is the average squared distance from the mean.
o Deviations squared=Variance
~ Step Four:
o Goal: to compute a measure of the “standard” (average) distance of the scores from the mean
o Variance measures the average squared distance from the mean—not quite our goal
~ Adjust for having squared all the differences by taking the square root of the variance
Standard Deviation = Variance
o Variance (in squared distance units) is not intuitively easy to grasp despite being a measure of
average squared distance of scores from the mean.
o Consequently, it is important to emphasize the need to take the square root of the variance to
return it to the same distance unit used in the original measurement procedure.
o So no we can speak to the Standard Deviation:
o This is the square root of the variance. So in order to calculate the SD you first have to calculate
the Variance. The SD provides us with the average distance from the mean.
Figure 4.2 Calculating Variance and Standard Deviation
~ So this diagram can be used to summarize this entire process.
~ Firstly you find deviation score for each score then to get rid of the + and – signs you square each
deviation score.
~ Then we want to find the average of the squared deviation, and this is by adding up those squared
values and then divide by the number of scores this will give you the variance.
~ And finally, to find the standard deviation, you take the square root of the variance.
~ So you will square root that variance score and that will give you the SD.

Measuring Variance and Standard Deviation for a Sample


~ Goal of inferential statistics:
o Draw general conclusions about population
o Based on limited information from a sample
~ Samples differ from the population
o Samples have less variability
o Computing the variance and standard deviation in the same way as for a population would
give a biased estimate of the population values
~ This podcast will focus on measuring variance and SD for a sample.
~ By now we should know the aim of inferential statistics is to draw conclusions about a population
based on limited info from a sample.
~ Therefore it is of utmost importance to make sure any sample is truly representative of a population.
The goal is also to find patterns and results in our data.
~ The amount of variability in the data influences of easy it is to pick up those patterns.
o High variability obscures patterns that would be visible in low variability samples.
~ Samples different from the population.
~ A sample statistic is said to be biased if it overestimates or underestimates the population. However it
is relatively predictable to pick up on bias in a sample and therefore it can be corrected.
~ So sample variablility isn’t a mistake it is a measure of how much a value is different from the “true”
value.
~ So lets say as an example we find that the true weight of a population is 90kgs.
~ You take a sample and find the mean weight is 91kgs.
~ The difference here is 1kg.
~ If you sample again you might get different mean weights of 89, 87.5 etc, so the difference is a
reflection of variability in your sample.
~ Relating to another example: the speedometer on your car.
~ Maybe your speedometer consistently shows speed that is 5km per hour slower than you are actually
going, it does not mean that the speedometer is useless, its simply that maybe your speedomter
needs an adjustment.
~ The main aim of that adjustment is then to make sure that the resulting value for the sample is
accurate and unbiased.
Formulas for Sample Variance and Standard Deviation
~ Sum of squares (SS) is computed as before
~ Formula for variance has n-1 rather than N in the denominator
~ Notation uses s instead of σ

~ When you are calculating variance for samples the only difference is that the denominator has this n-1
adjustment.
~ For samples, we take the sample size minus 1. the variance for a sample is equal to the sum of
squares divided by the sample size minus 1 and in order to get the SD we just take the square root of
the variance.
~ So when receiving any sample of data and you need to calculate variance and SD for a sample you
make that n-1 adjustment.
~ So why do we do this?....
Figure 4.4 Population of Adult Heights
~ The population of adult heights forms a normal distribution.
~ We know this because when we draw a line through the middle, it will be symmetrical.
~ If you select a sample from this population, you are most likely to obtain individuals who are near
average in height.
~ As a result, the scores in the sample will be less variable (spread out) than the scores in the
population.
~ So, the reason is that samples underestimate the true variability in the population just like we spoke
of in the first slide.
~ Here you can see the entire population and if we just sample 10 or 15 of these individuals we might
me lead to believe there’s less variability inherent in the distribution than really exists.
~ So, by adding that adjustment it then reduces the denominator values and inflates the SD slightly to
correct that initial underestimate.
~ So, let’s look at some examples to calculate this in the video next
Example: Calculating Sum of Squared Deviations (SS)
(å " ! )
SS=åx2 - $
10,7,6,10,6,15 n=6
10+7+6+10+6+15=54 åx=54
10 +7 +6 +10 +6 +15 =546
2 2 2 2 2 2
åx2=546
(%&)!
SS=546- '
=546-486
=60
Example: Calculating Sample Variance
((
s2 =$)* SS=60
n=6
'+
s2 =')*
'+
=%
s2 =12
Example: Calculating Standard Deviation
S=√12 s2 =12
= 3.46
Sample Variability and Degrees of Freedom
~ Population variance
o Mean is known
o Deviations are computed from a known mean
~ Sample variance as estimate of population
o Population mean is unknown
o Using sample mean restricts variability
~ Degrees of freedom
o Number of scores in sample that are independent and free to vary
o Degrees of freedom (df) = n – 1
~ So we know that with the variance for a population the mean is known, unlike with variance for a
sample where the mean of a population is unkown.
~ Hence we measure distance from the sample mean.
~ And we know that we must first compute the sample mean before we can begin to compute
deviation.
~ But calculating the value of M places a restriction on the variability of scores in a sample.
~ This can be demonstrated in the following table.
X A sample of n=3 scores with a mean of 5

2
9
------ (What is the third score?)
~ For example you have sample of n=3 scores and compute a mean of M=5.
~ The first two scores in the sample have no restrictions and we can see that the third score is
restricted, so what do we do?
~ We see that the third score must be 4. we say 4 because the entire sample has a Mean of 5 so for 3
scores to have a mean of 5 the total must be 15.
~ So get to 15 we can see that the first two scores added together gives us 11 therefore 4 is left to make
up 15, hence we say that 4 is the restricted value.
~ So the first two scores were free to have any value but the third score was dependent on the first
two.
~ So a the first sample of n-1 scores are free to vary but the final score is restrict therefore we say that
as a result, the sample is said to have n-1 degrees of freedom.
~ The degrees of freedom determine the number of scores in a sample that are free to vary.
Chapter 5: z-Scores: Location of Scores and Standardized Distributions
Introduction & Purpose of z-scores
~ Identify and describe location of every
score in the distribution
~ Standardize an entire distribution
~ Take different distributions and make them equivalent and comparable
~ In this chapter we will discuss z-scores and the location in a distribution.
~ So, we will learn how we can take any raw score of a test or an assessment for example and convert it
into a standardized score.
~ Okay, so when we talk about z-scores we talk about turning a raw score value into a z-score or a
standardized score.
~ Why?
~ Two purposes: when we convert a raw score into a standard score, we are able to tell the exact
location of an original score of the entire sample of people that took the test.
~ Now more importantly z-scores or standardized scores allow us to directly compare the results of
scores that come from two completely different distributions that have their own mean and own
standard deviation.
~ So, by taking those two raw scores and putting it on a scale we can see which is more or less
competitive than the other.
~ One practical example you can use to introduce z-scores and standardization involves baseball.
~ How could you compare the performance of a player in 1968 to one in 2000?
~ We know that scoring was much lower in 1968, so if we simply looked at their raw batting averages,
or home runs hit, the player from 1968 would likely appear to be much worse.
~ But if we standardize their scores—by comparing them to the mean batting average (or home runs
hit) in 1968 and 2000, we now have a common metric to compare the scores: by how well they did
relative to other players in that season.
Figure 5.1 Two Exam Score Distributions

~ Two distributions of exam scores.


~ For both distributions, μ = 70, but for one distribution, σ = 3 and for the other, σ = 12.
~ The relative position of X = 76 is very different for the two distributions.
z-Scores and Location in a Distribution
~ Exact location is described by z-score
o Sign tells whether score is located above or below the mean
o Number tells distance between score and mean in standard deviation units
~ So calculating z-scores is relatively simple.
~ So z-scores can either be positive or negative and this tells us that the result is either below or above
the average.
~ The number itself tells us how many how many sd units or brackets that score lies above or below
the mean.
~ For example: If I calculate a z-score of +2.0 that means that the score is 2 full SD brackets above the
mean.
~ Alternatively a score of -2.0 tells us that the score lies 2 full SD brackets below the mean.
Figure 5.2 Relationship Between z-Scores and Locations

~ Relationship between z-score values and locations in a population distribution.


~ Okay, so here is our z-distribution, the mean of the z-distribution is 0 and the SD is 1.
~ Those are the characteristics of z curve.
~ This is important to remember.
~ Right, so we will look at computing z-scores for sample and population in the next podcast.
Computing z-Scores for Samples and Populations
~ Populations are most common context for computing z-scores
~ It is possible to compute z-scores for samples
o Indicates relative position of score in sample
o Indicates distance from sample mean
~ Sample distribution can be transformed into z-scores
o Same shape as original distribution
o Same mean M and standard deviation s
~ In today’s podcast we will touch on computing z-scores for samples and populations.
~ So populations are the most common context for computing z-scores usually we want to see where
someone falls relative to the rest of the population however we can also compute z-scores for
samples.
~ So the definition and purpose of a z-score is the same for samples and populations provided that the
mean and sample mean and sample standard deviation are used to specify a z-score location.
~ So when computing z-scores for samples and populations, the X value is transformed into a z-score
so that:
~ The sign of the z-score indicates whether the X value is + or - so above or below the mean.
~ Now lets look at the equations for calculating the z-score for samples and populations before we
actually get into the way in which to calculate these scores…
Equation for z-Score for a Population

~ Numerator is a deviation score


~ Denominator expresses deviation in standard deviation units
~ Remember that z-scores identify a specific location of a score in terms of deviations from the mean
and relative to the standard deviation.
~ Okay, here is the equation for a z-score of a population.
~ So in order to take any raw score and turn it int a z-score, we simply take that score and subtract the
mean from that value and then we divide by the SD.
~ So essentially we look and say how different is that value (X) from the population mean (u) and we
divide by the SD.
~ Look at the video in the next slide for an example on how to calculate the z-score.
Calculating z-Score for a Population
~ Remember that z-scores identify a specific location of a score in terms of deviations from the mean
and relative to the standard deviation.
~ Okay, here is the equation for a z-score of a population.
~ So in order to take any raw score and turn it int a z-score, we simply take that score and subtract the
mean from that value and then we divide by the SD.
~ So essentially we look and say how different is that value (X) from the population mean (u) and we
divide by the SD.
~ Look at the video in the next slide for an example on how to calculate the z-score.
")µ
z= s
µ=100 s=10 x=130
*,+)*++
z= *+
,+
=*+
z=3.0
µ=60 s=9 x=65
'%)'+
z= -
%
z=-
z=0.55 / 0.56
Equation for z-Score for a Sample

~ Numerator is the score - mean


~ Denominator expresses standard deviation
~ So the z-score for a sample is expressed in this formula here where each X value in a sample can
be transformed into z-scores by using this formula.
~ So we say that z-score equals raw score value which we input as the X value minus the mean
then divided by the standard deviation.
~ Here for samples the standard deviation is symbolized with the letter s.
Calculating z-Score for a Sample
~ Remember that z-scores identify a specific location of a score in terms of deviations from the mean
and relative to the standard deviation.
~ Okay, here is the equation for a z-score of a population.
~ So in order to take any raw score and turn it int a z-score, we simply take that score and subtract the
mean from that value and then we divide by the SD.
~ So essentially we look and say how different is that value (X) from the population mean (u) and we
divide by the SD.
~ Look at the video in the next slide for an example on how to calculate the z-score.
").
z=
(
M=40 x=35 s=10
,%)&+
z= *+
)%
= *+
z = -0.50
M=60 x=48 s=9
&/)'+
z= -
)*0
= -
z = -1.33
Determining a Raw Score (X) from a z-Score for Populations
X -µ
z= so
s X = µ + zs
~ Algebraically solve for X to reveal that…
~ Raw score is simply the population mean plus (or minus if z is below the mean) z multiplied by
population standard deviation
~ So although the z-score equation works well for transforming X values into z-scores, it can be a bit
awkward when you are trying to work in the opposite direction by changing z-scores back into X
values.
~ So this is done by using the equation on the right. X= mean (population) plus z score multiplied by
the standard deviation.
~ Remember the z score describes exactly where the score is located by identifying the direction and
distance from the mean.
~ So lets look at some examples in the video on the next slide…
Examples
x=µ+zs
z=2.00 µ=60 s=9
=60+18(2.00x9)
x=78
Determining a Raw Score from z-Score for Samples
~ Similarly each z-score can be transformed back into a raw score with this formula here for samples.
~ We are looking for the X value so we say X equals the Mean + the z-score multiplied by the standard
deviation.
~ Lets look at some examples in the video on the next slide.
Examples
x=M+zs
M=40 z=1.00 s=10
=40+10
x=50
Using z-Scores to Standardize a Distribution
~ Now we spoke about this in the previous podcast and we know that every X value can be
transformed into a corresponding z-score.
~ Now more specifically, if every X value is transformed into to a z-score, then the distribution of the z-
score will have these characteristics:
~ shape: the distribution of z-scores will have the same as the original distribution of scores.
~ Every X value can be transformed to a z-score
~ Characteristics of z-score transformation:
o Same shape as original distribution
o Mean of z-score distribution is always 0
o Standard deviation is always 1.00
~ A z-score distribution is called a standardized distribution
~ So for example if the original distribution is positively skewed, the z-score distribution will also be
positively skewed.
~ So as I have mentioned before, if we have a z-score of +1.00 this means that it is above the mean by
1. likewise if its -1.00 it means that it is below the mean by 1.00.
~ 2. The mean: In a z-score distribution the mean will always be zero.
~ So all positive z-scores are above the mean of zero and all negative z-scores are below the mean of
zero.
~ 3. The standard deviation: the distribution of z-scores will then always have a SD of 1.
Figure 5.5 Transforming a Population of Scores
~ An entire population of scores is transformed into z-scores.
~ The transformation does not change the shape of the population, but the mean is transformed into a
value of 0 and the standard deviation is transformed to a value of 1.
Using z-Scores for Making Comparisons
~ All z-scores are comparable to each other
~ Scores from different distributions can be converted to z-scores
~ z-scores (standardized scores) allow the direct comparison of scores from two different distributions
because they have been converted to the same scale
~ Remember we said that we can compare z-scores to each other.
~ So whether you look at Karen’s Psychology test score and compare it with her English test score, we
can directly compare the scores because they have been converted into a z-score.
Chapter 6: Probability
The Unit Normal Table
~ The proportion for only a few z-scores can be shown graphically
~ The complete listing of z-scores and proportions is provided in the unit normal table
~ Unit Normal Table is provided in Appendix B, Table B.1
~ This podcast will demonstrate how to use the Unit Normal Table.
~ So this table lists the proportions of the normal distribution for a full range of possible z-scores.
~ The table that I will show you next is just a portion of the complete table but you will be able to find
the compete table in you textbook under the appendices.
Figure 6.6 Portion of the Unit Normal Table

~ So this Unit Normal table we can see at the top of each column A, B, C, and D.
~ So remember what we learnt in the previous podcasts that transforming an x-value we use the
equation z= x-u divided by SD for populations and for samples we use z=x-m divided by s.
~ Son once you transform your raw score into the z-score you enter the z-score into the table under
column A, but this is given here already in the table.
~ Please also note that the z-score is inputted as 2 digits right at the decimal, so adhere to the standard
rounding rules.
~ The next column B represents Proportion in the body, and over here we can see an illustration of
what it looks like.
~ So if we are interested in the area or probability of proportion in a distribution below a positive z-
score, again this would be positive because it is above the mean, we would sketch the distribution like
this and we would transform the x-value into a z-score and then report the are in column B which
represents the proportion in the body which will always be greater that point 5 or 50%.
~ Here we can mirror the negative z-score
1)µ
z= s
~ So again I say a normal distribution is symmetrical right, so if we look at a negatively skewed
distribution where we could have a negative z-score so the value is below the mean.
~ Now in the next Column C, we would be interested in the area above a positive z-score.
~ So a value above a positive z-score which we may tend to see as the smaller are in the distribution.
~ So because the distribution is symmetrical we can say the same for the other side in terms of below a
negative z-score, again the proportion or probability of a score that is less than the mean.
1)2
z= (

~ Finally we have column D, which is refereed to as the proportion between the mean and z.
~ So we may be interested between the proportion and probability of score between the mean and a z-
score.
~ So here in the distribution it pertains to a positive z-score and we can consider the same, the area
between the mean and the z for a negative z-score.

~ Okay, so in the next podcast I will show you how to use the unit normal table to find proportions and
probabilities for a specific z-score.
Probability/Proportion & z-scores
~ Unit normal table lists relationships between z-score locations and proportions in a normal
distribution.
~ If you know the z-score, you can look up
the corresponding proportion.
~ If you know the proportion, you can use the table to find a specific z-score location.
~ Probability is equivalent to proportion.
~ In this podcast I will show you how to find the Proportions / probabilities for a specific z-score and Z-
Score locations that correspond to specific proportions.
~ There are a few steps to follow:
~ Firstly, to find proportions or probabilities for a specific z-score value you always:
~ Sketch the distribution and shade the are of interest.
~ Transform X values into z-scores
~ Enter the Unit Normal Table using column A and reference the appropriate proportion (Body, Tail, or
Area between Mean and z) according to the sketch drawn in the first step.
~ Remember it is vital to first sketch the distribution. If you dont you may make careless mistakes.
~ Now lets look at an example… In the next slide
~ Remember our equations for z-score…
Example
~ A sample is normally distributed with a mean of u=45 and a standard deviation of o=4.
~ What is the probability of randomly selecting a score that is greater than 43?
Refer to the Table

Z-Score locations that corresponds to specific proportions


Example
~ What z-score represents the top 10 % from the remainder of the distribution?
µ=24.3 min
s=10 min
10%
10%/100 =0.10

x=µ+s(z)
=24.3+10(1.28)
=24.3+12.8
=37.1min. ->the 10%
The Unit Normal Table to Locate the z-Score
Calculating the X value corresponding to proportions or probability
~ The probabilities given in the unit normal table will be accurate only for normally distributed scores so
the shape of the distribution should be verified before using it.
~ For normally distributed scores
o Transform the X scores (values) into z-scores
o Look up the proportions corresponding to the z-score values

You might also like