Basic Statistic
Dr Mohd Radzi Tarmizi Bin A Halim
Who work with statistic
• Statistic practitioner = person who use statistical technique properly
• Statistician = individual who work with the mathematics of statistics.
Their work involved research that develop technique and concept
that in the future may help the statistic practitioner
• Psychologist? = Statistic practitioner, using applied statistic (Cohen
2013)
What is statistic
• A way to extract information from set of data (Warrack 2003)
• Is a set of tools and techniques that is used for describing, organizing, &
interpreting from data (Salkind 2000)
• Statistic can be visualize and be explain – in graph, chart
• Consist of a body of methods for collecting and analyzing data
• refers to a collection of numerical facts, such as a set of performance measures
for a baseball team (e.g.,batting averages of the players) or the results of the
latest U.S. census (e.g., the average size of households in each state of the United
States) (Cohen 2013)
• statistics refers to a branch of mathematics that is concerned with methods for
understanding and summarizing collections of numbers. So the answer to “What
is statistics?” is that it is a set of methods for dealing with numerical facts.
Psychologists, like other scientists, refer to numerical facts as data.
Branch of statistic
• Easier
• Use to explain only
Desciptive • Eg: Percentage, Mean, Frequency,
Median etc
Statistic
• More complicated
• Use to make generalization &
Inferential hypothesis testing
• Eg: Correlation, Regression, t-test,
ANOVA etc
When to use each branch?
• Both are important & rely on each other
• Both use in thesis, research paper & other real life application such as
in predicting most suitable candidate for job recruitment
• Depend on your research question & hypothesis (Salkind 2000)
Statistic & Research
• Statistic is used in many psychological research
• For example, what is the effect of milk to sleep
• Descriptive statistic will explain the duration of sleep & how many
milk are consumed by respondent (frequency, mean, mode, median)
• While inferential statistic will explain if milk has influence on sleep or
not (using regression) and whether there is different in sleep onset
with those who consume milk and those who don’t (t-test)
Variable & Constants
• Something that different or something that you manipulate in
research, its called variable
• Something that is same or you doing nothing in research, its called
constant
• Setting variable or constant is depend on your research question and
hypothesis Independent Independent Dependant
• For eg: Variable Variable Variable
Milk (ml) Type of Mattress Duration of Sleep
(Constant) (Variable) (hours) (Variable)
250 Hard 3
250 Soft 5
250 Super Soft 8
Scales of Measurement
• Measurement is a system for assigning numerical values to
observations in a consistent and reproducible way
• For eg, to measure milk volume, we use mililiters (ml), to measure
time, we use seconds, minutes or hours and to measure heart rate,
we use beat per minutes.
• Stevens (1946) introduce four scale systems:
• Nominal Scales
• Ordinal Scales
• Interval Scales
• Ratio Scales
Nominal (1 st Level)
• Nominal refers to the fact that the values are simply named, rather than
assigned numbers
• If numbers are assigned to the values of a nominal scale, they are assigned
arbitrarily and therefore cannot be used for mathematical operations. (eg:
We may assigned Male as 0 or Female as 1).
• Used to something that makes no sense mathematically (eg: DSM-5 is not
severe compare to DSM-4 because the number is increased)
• Also known as Categorical data/ qualitative data
• Also referred to as a qualitative level of measurement because each level
has a different quality and therefore cannot be compared with other levels
with respect to quantity
• Nominal: Department: Faculty of Science, Faculty of Medicine, Faculty of
Maritime etc
• Binary (Male or Female; True or False; Live or Dead)
Ordinal Scales (2 nd level)
• Also grouped in Categorical data/ qualitative data
• A quantitative level of measurement, different value of scale can be place in order
• These designation have a meaningful order, that why its called ordinal scale
• Often level of an ordinal scale are given numbers. For eg: ranking handwriting of student at Bad
(0), Fair (1), Good (2), Excellent (3). Or Ranking the marathon participant winner to Gold (1), Silver
(2) and Bronce
• However, the rankings cannot be treated as real numbers; that is, it cannot be assumed that the
third-ranked gymnast is midway between the second and the fourth.
• In fact, it could be the case that the number 2 gymnast is much better than either number 3 or 4,
and that number 3 is only slightly better than number 4 (as shown in Figure 1.1). Although the
average of the numbers 2 and 4 is 3, the average of the abilities of the number 2 and 4 gymnasts
is not equivalent to the abilities of gymnast number 3.
• Other example, score of depression between 20-30 is not same difference between 40-50
• Likert scale is listed under ordinal scales
Interval Scale (3 rd level)
• Beyond the ordinal scale
• Is a continuous scale (can be 1, 2, 3…or 1.1, 1.2..1.9, 2.0…)
• Has an equal distance or spaces between two variables
• Has unit of measurement (eg: Length in meter, Volume in liter,
Temperature in Celcius)
• Length of 1cm is same whether we measure from 0cm to 1cm or started at
4cm to 5cm
• Strongly Agree (5) to Strongly Disagree (0)
• Lack of true Zero (0), means 0 has its own meaning. Or, 0 represent of the
presence of the property being measured. 0 Kelvin = -273.15 Celcius
• Absolute/True Zero = 0 point represent the absence of the property being
measured ($0 = no money, no behavior, none correct, no agreement)
Ratio Scale (4 th level)
• Also a continuous scale
• when your measurement scale tells you that you now have twice as
many units of the variable as before, you really do have twice as much
of the variable.
• All ratio scales have the interval property
• IQ score of 150 is not exactly twice as intelligent as someone who
scores 75.
Parametric Vs Non Parametric Statistic
• Nearly all common statistical procedures are just as valid for interval
& ratio scales
• Statistical procedures based on distributions and their parameters are
called parametric statistics
• Interval & ratio scale often to use parametric statistic, because it was
designed to be use with interval/ratio scale
• Parametric statistic also can be use when the data meet with
distributional assumption (the data is normally distributed)
• If the data is not normally distributed or the data gathered from
nominal or ordinal scales, then non-parametric statistic will be use
Likert Scale and Measurement Controversy
• Most psychological research need respondent to response to a statement by indicating
their degree of agreement on Likert Scale, named after its creator Rensis Likert (1932)
• Common Likert scale is 5 or 7 point of agreement (Strongly agree to Strongly Disagree).
But the difference of 1 (Strongly disagree) and 2 (Disagree) is not equal to 4 (Agree) and
5 (Strongly Agree).
• This Likert scale possess ordinal property & Stevens (1951) says it is not permissible to
perform mathematical operations, like average on numbers derived from ordinal scales.
• However, statistician like Vellemen & Wilkinson (1993) said what most matter in
determining which type of statistic can be validly applied to your data is the type of
questions you are asking of your data and what are is the goal of your research
• Norman (2010) said that there is no problem in applying parametric statistic to ordinal
data because the procedures are robust with respect to the interval scale assumption-
that is a lack of equality of intervals by itself has little impact on the final statistical
conclusion.
Continuous Versus Discrete Variables
• Continuous variable has infinite numbers
• Eg: Height, Weight
• 1.0… 1.1… 1.199… 2.0…
• Discrete variable has finite number with no intermediate possible
value
• Number of people 1, 2, 3 etc..
• Number of family
• Number of education level
• Both continuous & discrete variable does not affect the inferential
statistic, as long as the variable is measured on a quantitative scale
Scales Versus Variables Versus Underlying Constructs
• Variable can be measure using different scale. Eg: outside
temperature can be measure as an ordinal (cold, hot, hottest),
interval (Celsius or Fahrenheit) or ratio (Kelvin)
• Choose a right scale to operationalizing (define how, what and when
the measuring procedure) a variable so that it can be replicate by
other researcher.
• To measure student weight, we can use weight scaler by standing on it after
shower
• Underlying construct is what we can understood intuitively but hard
to define. Eg: how we define generosity? By money he donate or by
the total of charity per year or how many families he has help
Independent Versus Dependent Variables
• Independent variable – variable that we manipulate and it can give
effect on dependent variable
• Dependent variable – variable that affected by the manipulation of
independent variable
• Sugar intake vs sugar blood level
• Personality vs job performance
• Time of studying vs % mark
• Depression vs suicide