0% found this document useful (0 votes)
31 viewsPsych Stats Sum
Uploaded by
Sailor MoonCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
31 viewsPsych Stats Sum
Uploaded by
Sailor MoonCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 10
Peychology
© Scientific study of behavior rd mental processes
> Scientific -there isa method ar process involved
© Goals:
1 BBB who: happening?
2. RBI - wy someshing appening?
when wilt happen agin?
CConvertesto Statistics
© Always stars with 2 phenomenon, an experienc, 2 situation
© Numbers hold mearing — convey astory
1 Facts and figures, set of mathematical procedures that share a story and give meaning
Schavor isexplened and predicted through educated guesses (hypothesis) ina precise are
structured manner
Untot Observation
© Usitof ness. — ES os SETI. o ps)
Hypothesis Testing
© Statements of relationship between two or more variables
‘Variables:
‘© Characteristic that eitfer fom on®inclvidual to another, or from one polntin time to another
© Natal variables vary (constant: gender)
(BEBNSEEREE - 250s, invivences, or has an impact on another variable;
‘ontroleg by an experimenter
(DEBRIS Variable ffected,infuenced,resuitart ofthe independent
vatiable
‘= Examples Iv: residence type DV: friendliness
We testthe hypothesis in order to know ifsomethingis true fake, orbaseess fad not just
because we believe it)
= Not biasec by cur preconceptions ane personal belief
= We shouldbe abe to criticize assumptions, ideas, and concepts of behavior ‘Stages Sail Rasoareb
Problem reduced toa testable hypothasis
Developing instruments (vay, alibi, psychometi propertis)
Data collection how to collec? Depends on your problem/study/target),
Data anaiyed to relative hypothesis — guide in statistical analsis
Results of analysis siterpreted and shares
Euncions of Statistics
BEBE: wn st do you see inthe date?
‘© Setof techniques forthe reduction ofthe quantitative data (numbers to smaller number or
"more convenient and easly communicated terms
‘© What can the dats tll me about the participants?
fe Setof decision making techniques aiding researchers in making inferences
‘© Can elpthe researcher generalize the findings from small samples to larger groups of people
(oepulations)
Statistically significant’: probabllty that it occured by chance is small?
Make canlusions assed an your study
‘© Statics wil pve youalevel of confidence ta make decisions netween the sample and the
population and about the hypothesis being tested
‘Measurement
Nature of Data
‘© Some canbe well defined and easly quartifable (ex: eye color, height, weight, shoe size)
© Some are not well-defined (ex intelgence, humor, anxiety, sadness)
This s when researchers ue SSE :0 explain the constuc (ex anseny
[toe general)..alpitations [operational definition])
‘© Countable, onlyafinte number of values possible, cannot be subdvied meaningfully
Nominal and ordinal
‘Continuous Data
‘© Measured ona continuum or scale can take ary numeric value, can be divided into smaller
fee subdivisions
Ratio and intel
‘©The term continuous apis tothe variable being measured, not the scores abained
© Continuous ata nas RBI - ea! iis ae haltways berween adjacent categories
Scales of Measurement
"© Provides name or etegary foreach abject (the mumbers serve as labels)
‘© Every case must be mutually exclusive (does net overlap) and exhaustive (all data canbe put
Into the categories) ‘© Quality and quantity isnot implied
Classify or organize cata
Ordinal
‘© Rank the data but cannot tel how much is the diffrence from one rank to another
Rank or order
f¢ Measurements consist of equal intervals between values (ex temperature, |, time of the day
inure)
Assign ascore
‘ualtes oF interval scales pus there fsa true zero point [time in seconds, welt yeas of
experience)
© Assign score
Bxample:
oppiness in my chosen organization
Nominal: Are you happy with your chosen ong?[ yest] no
‘Orinal: How happy are you withthe org?[ Disappointed Content ]Satisied | Happy
Interval Ratio: | am happy with my org
[ )2-Stronsly disagree
1 12-Disageee
( 13-Agree
[ )4-Stronaly Agree
"Your numbers have meaning, & can produce total points
Youllbe able to wtzestatstial analy,
Paramevic ang Non-parametric
‘© Twobroad classifications of statistical procedures
© _Basicdistinction
‘© information about the population is completely known
‘+ asad on assumptions about the lstibution of underlying population fom which yoursarale
was taken (normal distribution)
1 Test statistics s based on aistioution
Mote powerful
Non-parametric
© Noinformation about he population s avaiable
© Donat rely on assumptions about the shape or parameters ofthe underlying population
distribution
Test statistles is arbitrary
Less powertl Chapter?
‘Organizing the Data
Descriotive Statistics
‘© Youll always se his king, even in inferential satistes
© ERIUBERISERIREESINERRD — teastorms ra data into mesninaul ane orgonited set
of measures used to test hypothesis
‘Involves diferent ways or techniques of tabulatng and depicting collections of deta
‘© ftlonale the urnan ming isa limited information processor
© Before we are able to comprehend data and see meaning behind the figures, data must
undergo a process a orgaizing (2),
‘© Whore the datas in a particular catagory table format
VARIABLE: Test Score FREQUENCY: # of Respondents
‘36-100 4
9195 5
'¢ When the distributions among groupe are equal in sample size. You can use the frequency to
‘make comparisons between groups rarely hapoens()
‘© Standardizing forse and comparing clstributions especialy of hose of unequal sample size
P=f/N = Compares the umber of cases ina given category with the total size ofthe distribution
%4=11N* 100 — Frequency of occurrence ofa category per 100
— Multiplying the proportion by 100
‘© Categories nominalevel dstributions do nat have tobe in order
18 Usting of categories must e arranged in order (usually highest tofowest) allowing readablity
and logical sens (Ordinal and interval) Some scores are widespread making he frequency distribution long and dificult to ead
Solution: Condense separate scores
Each category or group in grouped frequency distributions known 2s SER, nese
ss determined by the number of score values in it
2 here's: TTT
How to choose inten?
From 2inteninleta as mony 35 29
‘Te intervale must show pattern
Baseit onthe date set, objectives ofthe research
Make ita whole number to make the calculations simple
[Make the lowest score ina classinterval multiple ofits sie
'¢ Misclemost score vale inthe class interval
©) in-lowestvalue- highest value
2
‘© Used when locating the postion of one case to overall roup performance
‘© Symbol: of for cumulative frequency and for cumulative percentage
(© Aaddthe frequency of a category tothe total frequency for all categories below it
+ SOS c/n" 00
crassa
"e Tobie that presents the distribution - frequencies and percentages. of one variable (usual the
dependent variable) across the categories of one or more addtional variables (usualy the
Independent variables)
‘© Series of frequency distributions atached together to make one table
‘© Frequency cstributions ofeach variable separately can be found along the margins of
two-way cross tabulation —|
= (200) {otal
1© Dividing each Frequency bythe total sample size
"= Row %= (100) /Nrow
'© Dividethe frequencies in each row by the numberof casesin that row
"© Col = [100 }1/Neslumn
‘© Dividing each frequency by the number ofasesin that columa 1. ftheindepandentvaribleis on the rove, use row percent
2. Ifthe independent variables onthe columns, use column percent
13 Ifthere sno clea-cutindependent variable, use total row, or column percents, whichever is
‘most meaningful forthe particular research focus
‘TWO DECIMAL POINTS ONLY
‘Graphic Presentations
Some people may get scared of aang string of numbersin table form
2 Intimidating
> How do tead those? How wil! interpret i
(© Convert the numbers to visual pletures and it becomes less scary
9 Itbecomesseexplanatory inane glance
Pie Charts
‘Creuar graph whose ses add up to 100%
Showcasing diferences in equencies/percentages among categories
Nominal variables
To highlight an aspect ofa chart, we can explode -move slighty outware
Disadvantages:
2 Distinction benween colors if there are many variables
2 leslie decreases in visual size as more variables are used
> Limited variables in splay
‘Bar Graph
Values plotted over x and axis ina graph with plas to show numerical score
‘Accommodate any numberof categories at any level of measurement
{Used 10 display the frequency or percentage distribution ofa discrete variable
“Theres space between the bars to emphasize differentness
‘Histogram
'¢ Used to display interval levels of measurement
‘©The bars ate joined to emphasize continuity ofthe points long ascale
‘© Number rangesin the sais
"© Shows the continuity along scale atherthan the citferences
Useful showing ordinal or nterval/eaio data
1¢ Points placed over score values or midpoint along the axes and are connected with a straight
line, whichis dropped ine charts
‘© Same principles frequency polygon except end points donot touch the bases
(© Trends are commonly depicted usingline charts
Ts:
1 Putthe legende at the anes ensure the readers know what varabes those values depict
© Use diferent colors
Shape and Distribution
‘© Frequency polygons show distribution of scores or values
(© Tals you where the scores grouped, where isit highest
‘Symmetrical
‘© There are same dstroutions to either side of the curve la polygon, bath ef and right (low and
ign)
© bell curve
1 Skewed -more cases in one direction than the other
© When skewness ens
2 The scores pile upin one direction
© There willbe a pronounced tail
Negative Skewes Normal(no skew) Positively Stewed