Psychological Statistics - Lecture
INTRODUCTION Datum (singular) - a single measurement or
Psychology observation andis commonly called a score or
- scientific study of human behaviour and mental raw score
processes Data Set
- Describe, Predict, Explain, Control - compilation of data
Statistics - in a file
- method of pursuing truth - collection of measurements or observations
- Statistical method
- Branch of mathematics that deals with After we gather the data, we analyze
organization, summarization, analysis, and
interpretation, of group of numbers Descriptive Statistics
- describe the data
Population - Statistical procedure used to summarize,
- set of all individuals of interest in a particular organize and simplify data
study
Inferential Statistics
Parameter - to generalize the truth
- a value, usually a numerical value, used to - consists of techniques that allow us to study
describe a population samples and then make generalizations about
Sample the populations from which they were selected
- a set of individuals selected from a population Sampling Error
- Intended to represent the population - the naturally occurring discrepancy, or error,
Statistics that exist between a sample statistic and the
- a value, usually a numerical value, used to corresponding population parameter
describe a sample VARIABLES AND MEASUREMENTS
Discrete Variable
Experimental Quantitative Qualitative - whole number
Sample - acquired through counting
Subjects Respondents Participants
- consists of separate, indivisible categories
Examples: number of students, number of correct
Variables answer
- characteristics or condition that changes or has Continuous Variable
different values for different individuals - have decimal numbers
- the one that we measure - acquired through measuring
- nagbabago - there are an infinite number of possible values
Values that fall between any two observed values
- to describe the variable Examples: speed, height, weight, temperature
- category or number Dichotomous Variable
- possible number or category that a score can - one that takes on one of only two possible
have values when observed or measured
Example: 0-20 in a Stress Scale, Male or Female Artificial Dichotomous - derived from scores
Score (may pinang gagalingan)
- results of the survey Example: failed or passed
- raw score True Dichotomous - naturally occurring
- a particular person’s value on a variable Example: Female or Male, Yes or No, True or
False, Heads or Tails
Data (plural)
- general Level of Measurements
- compilation of score (raw score) - another way of classifying things
- measurements or observation - a.k.a. scales of measurements
- classification that describes the nature of Survey Research
information within the values assigned to - common type of descriptive research
variables - useful way of obtaining data about people’s
Properties: opinions, attitudes, preferences, and
1. Magnitude - nagsasabi gano kalaki yung experiences that are hard to observe directly
value - may be obtained through questionnaires
2. Equal interval - equal interval alam mo na (close-ended) and interviews (open-ended)
yun Content Analysis (quantifiable)
3. Absolute zero - edi may zero - hahanapin yung thought or theme
nungsinabi
• Nominal
- a.k.a. categorical variables Relationship Between Variables
- categorization, classification, naming - two or more variables observed and
- walang magnitude, walang value or measured
pinagkukumpara - Used to determine what type of
- tinatanong ka lang kung ano yung kung ano relationship exist
yun Data Structure 2: One Group with Two
Examples: Sex, Nationality, Religion, Civil, Variables Measured for Each Individual
Status, What is you address?, What is your - one group of participants
jersey number? - goal is to describe the type and magnitude of
the relationship
• Ordinal
- a.k.a. rank-order variable - non-experimental
- order
- may magnitude pero walang equal interval Correlational Method
Examples: clothing size (S, M, L,), class rank, - two different variables are observed
grades (A-F), level of self-confidence (low, - To determine whether there is a relationship
average, high), likert scale between them
1. Positive Relationship (the same direction)
• Interval Kung mataas yung isang variable, mataas din
- numerical yung isa
- have equal interval but no absolute zero Kung mababa yung isang variable, mababa din
- rating difference yung isa
Example: temperature (Fahrenheit and 2. Negative Relationship (inverse)
Celsius), IQ, stress, pH, SAT score Kung mataas yung isang variable, mababa
• Ratio yung isa
- numerical Kung mababa yung isang variable, mataas
- have an absolute zero naman yung isa
Example: time to complete a task, number of Correlational Method Limitations
correct answers, weight or height gain in the - basta relationship padin naman siya
past 6 months, temperature (Kelvin) - not sure if may causation
DATA STRUCTURES AND METHODS - does not provide an explanation for the
relationship
Data Structure 1: One Group with One or - only demonstrate the existence of the
More Separate Variables Measured for Each relationship
Individual
- one or more variables measured per individual Data Structure 3: Comparing Two or More
- to describe one or more variables Groups of Scores
- variables are describes by descriptive statistics - one variable defined the groups
- scores are measured on second variable
Descriptive Research - both experimental and non-experimental
- a.k.a. descriptive research strategy
- involves measuring one or more separate Experimental Method
variables for each individual with the intent of - causation
simply describing the individual variables
- one or more variable is manipulated while Longitudinal Design
other variable is observed and measured - not ideal since long for college research
- aims to establish a cause-effect relationship - a method in which the same group of subjects
between the two variables and attempts to is followed and measured at different points in
control all other variables to prevent them from time
influencing the results - looks for changes across or over time
Extraneous Variable DESCRIPTIVE STATISTICS: FREQUENCY
- hindi siya part ng study but can have an DISTRIBUTION AND WAYS OF
impact sa IV and DV PRESENTING DATA
Hindi ginagamit ang term na IV and DV sa Frequency Distribution
correlational research - an organized tabulation of the number of
Independent Variable individual located in each category on the scale
- variable that is being manipulated of measurement
Dependent Variable Frequency Tables
- variable that is being observed or measured - ordered listing of number of individuals/
subjects/respondents having each of the
Experimental Condition different values for a particular variable
- a condition in an experiment wherein the
subjects will receive the experimental Proportion
treatment - measures the fraction of the total group that is
- compare group associated with each score
Control Condition Percentage
- a condition in an experiment wherein the - an amount of something often expressed as a
subjects do not receive the experimental number out of 100
treatment Grouped Frequency Table
Quasi - Experimental Design - frequency table in which the number of
- parang individuals is given for each interval of values
- seems like real experiments but they lack one Interval - range of values in a grouped
or more of its essential elements frequency table that are grouped together
- hindi kontrolado ng extraneous variable WAYS OF PRESENTING DATA
Nonequivalent Groups Design Histogram
- Example of quasi-experimental design - barlike graph pero ano dikit dikit yung bar
- a design in which the researcher compares - hindi pede sa categorical
the effect of different treatment conditions
on pre-existing groups of participants
Pretest/Post test Design
- used to assess whether the occurrence of an
event alters behavior or scores from
measurements made before and after the event
- hindi kontrolado ng researchers and mga
ginagawa ng participants after the experiment
or yung ginagawa nila sainterval nung pretest Frequency Polygon
and nung post test - parang line graph pero naka sarado
Ex-Post Facto Study - used for continuous data
- Postal comparative
- systematically examines the effects of pre-
existing subject characteristics by forming
groups based on these naturally occurring
differences between subjects
Column Chart - with equal interval variables: continuous data/
- categorical variables and interval/ratio
- Parang histogram pero magkakahiwalay na Weighted Mean
yung bar nito - average in which each observation in the
dataset is assigned or multiplied by a weight
before summing to a single average value
Median (Mdn)
- middle score when all the scores in a
distribution are arranged from lowest to highest
- rank-ordered variables
Outliers - score with an extreme (very high or
very low) in relation to the other scores in the
distribution
Mode
Bar Graph - value with the greatest frequency in the
- categorical distribution
- pagilid yung bar niya hindi patayo - categorical variables
- nasa baba naman yung frequency nito, syempre
naging pagilid na nga eehh SHAPES OF A FREQUENCY
DISTRIBUTION
Unimodal Distribution
- isa ang pinaka mataas (larger frequency)
Line Graph
- relationship with time only Bimodal Distribution
- Kung ang frequency polygon nakasara, dito - dalawa ang mataas (large frequency)
hindi
Multimodal Distribution
- madami ang mataas (high frequencies)
DESCRIPTIVE STATISTICS: MEASURES
OF CENTRAL TENDENCY AND SHAPE OF
DISTRIBUTIONS
Central Tendency
- typical or most representative value of a group
of scores
- mean, median, mode Rectangular Distribution
Mean (M) - approximately the same frequency
- arithmetic average of a group of scores
- sum of scores divided by the number of scores
- the balance point of a distribution
- commonly used in quantitative research
Symmetrical Distribution 4. Kurtosis
- kung ano yung nasa kabila yun din yung nasa - a frequency distribution deviates from a
kabila (like a mirror ganurn) normal curve in terms of whether its curve
- rare lang siya in the middle is more peaked or flat than
the normal curve
• Leptokurtic - the scores are concentrated
towards the mean
• Mesokurtic - normal curve
• Platykurtic - the scores have an extremely
large deviation from the mean
Skewed Distribution
- distribution in which the scores are pileup on
one side of the middle and are spread out on
the other side
- distribution that is not symmetrical
1. Positively Skewed Distribution
- papunta ang buntot sa right
Floor Effect - situation in which many scores
pileup at the low end of a distribution
DESCRIPTIVE STATISTICS: MEASURE OF
VARIABILITY
Variability
- vary, nagiiba iba
- provides quantitative measure of the difference
between scores in a distribution
- describes the degree to which the score are
spread out or clustered together
2. Negatively Skewed Distribution - represent how ka dispersed and data or yung
- papunta ang buntot sa left entire distribution
Ceiling Effect - situation in which many Low Variability - magkakadikit ang value ng
scores pileup at the high end of a distribution respondents
No Variability - magkakatulad ang value ng
respondents
High Variability - magkakalayo ang value ng
respondents
Range
- a distance covered by the scores in a
distribution from the smallest score to the
3. Normal Curve largest score
- edi normal nasa gitna lang - unreliable kasi merong outliers
- bell-shaped frequency distribution that is Variance
symmetrical and unimodal - meausre of how spread out a set of scores are
Deviation or Deviation Score - the difference
between a score and the mean2
Standard Deviation
- square root if variance
- most common descriptive statistics for
variation
- approximately the average amount that scores
in a distribution vary from the mean
Standard Deviation close to 0 - data points are
close to the mean
High or Low Standard Deviation - data points
are respectively above or below the mean
Definitional Formula
- equation for a statistical procedure directly
showing the meaning of the procedure
- Kung anong meaning yun yung formula
Computational Formula
- equation mathematically equivalent to the
definitional formula
- easier to use for figuring by hand
- does not directly show the meaning of the
procedure
VARIANCE AND STANDARD DEVIATION
FOR SAMPLES
Biased Statistics
- if the average value of the statistics either
underestimates or overestimates the
corresponding population parameter
Unbiased Statistics
- if the average value of the statistic is equal to
the population parameter
How to be unbiased?
1. Random selecting
2. Right sample size
3. Right statistical treatment
Median Absolute Deviation (MAD)
- a robust measure of how spread out a set of
data is
- good for assessing dispersion of data with
extreme values and/or non-normal shape of
distribution (outliers)
- doesn’t mind yung outliers yet nakakapagbigay
ng accurate na result