Class Notes (835,137)
Canada (508,956)
Psychology (3,518)
PSY201H1 (61)
Lecture 2

Lecture 2 - Frequency Distributions, Tables vs. Graphs

5 Pages
128 Views
Unlock Document

Department
Psychology
Course
PSY201H1
Professor
Alison Luby
Semester
Fall

Description
PSY201H1F L2 Sept 21, 2011 ○ List of x values is way too long for simple presentation of data Frequency Distributions, Tables vs. ○  Use a Grouped Freq Distr Table Graphs  X column lists class intervals: grps of • Frequency distribution: Organized scores ○ Guidelines: tabulation of entire set of scores allowing to see big picture in data at a glance  Select intervals so have about 10 ○ Shows where each indiv is located relative class intervals to others in the distr  Class interval width should be ○ Shows whether scores are high or low, simple number (2, 5, 10, multiples of concentrated in centre or spaced apart 5 or 10)  Bottom score in each class • Places raw scores in order from highest to lowest interval should b multiple of width ○ For nominal, any order • Ex. If width is 5 then bottom score • Group together everyone w same scores  is multiple of 5: 5,10,15,20 get sense of patterns  Intervals have same width, cover • Chart, table, graph, always presents 2 complete range of scores, each score belongs in 1 interval elements: ○ Cats that make up the original • Ex. Time to complete puzzle: 23  213 range measurement scale (128,99,213,117,83,201,79,156,88,91,23,65,1 ○ Record of the frequency (# of indiv in each 41,43,52,81,126,113,166,142) cat) ○ X column (top to bottom): 200-219, 180- 199,160-179,140-159,120-139,100- Freq Distr Tables 119,80-99,60-79,40-59,20-39 • Consist of at least 2 columns: ○ X column (raw scores) measurement cats ○ Start at multiple of width, don’t leave ○ F (frequency) column out intervals even if f=0 ○ Ex. Number of cars ppl have: ○ Less groups  more info you lose, less you know about your data 1,2,1,0,3,4,0,1,1,1,2,2,3,2,3,2,1,4,0,0 • Continuous var have an infinite number X F 4 2 of possible values, so in 200-219 a could 3 3 have X = 201.9 2 5 • Measurements indicate an interval not a 1 6 single pt ○ Ex. If X = 10 then the score was btwn 9.5 0 4 & 10.5 ○ Tally freqs for each X  How often each X occurs • Grouped Freq Distr Table:Values listed in  Σf = N the intervals are called Apparent Limits  Ex. Above: Σ f = 20 of the interval • But upper & lower boundaries involve • If you are asked for ΣX, do not add up the Real Limits values in the X column so don’t use the freq ○ Ex. For X=20-39, lower & upper real limits dirt table ○ Add up all of the raw scores for X instead are 19.5 & 39.5, apparent limits are 20 & ○ ΣX = ΣfX 39 ○ Or just add fX ex. 2(4) + 3(3) … = ΣfX Freq Distr Graphs • Pic (graphical representation) of info in a freq ○ ΣX = 33 distr table ○ Use the same method to calculate ΣX 2 • Calculate proportions&percentages to • All graphs have 2 perpendicular lines called describe distr axes which typically meet at 0 ○ Proportion (aka Relative Freq) ○ X axis = Abscissa: Lists the measurement scale (the set of X values) measures fraction of total grp ○ Y axis = Ordinate: Lists the freqs associated w each score ○ General rule: Y axis height = approx 2/3  p= f/N  Σp = 1.00 - 3/4 length of X axis ○ Percentage = p(100) = f/N x (100) • When score cats consist of numerical scores from interval or ratio scale, graph  Σpercentage column = 100% should be histogram or polygon • If many X values: ○ Don’t want many rows • When the scores r measured on a nominal or • Polygons: ordinal scale, the freq distr should be ○ Centre a dot above each score so displayed in a bar graph height of dot correspond to freq • Histograms ○ Join dots w straight lines ○ List cats of measurement along X axis ○ Additional line drawn @ each end to ○ Draw a bar about each X value so: bring graph back to 0 freq  Height of bar corresponds to freq  Usually reach the x axis at a pt 1 cat for that cat below lowest score on left & 1 cat  Width of bar extends to real limits above highest score on right of the cat ○ For Grouped Distr, position each dot •  No space btwn bars directly above midpt of the Class  When data grouped into class Interval intervals, draw bar above each interval  Find midpt by avging highest & lowest so width of bar extends to real limits of score for the interval the interval ○ Similar to freq histogram ○ Ex. Freq of quiz scores – Use Reg Freq  Appropriate for plotting freq of Distr Table continuous vars  Communicates same info  Distr shape may be easier to see using polygon ○ Using Reg Freq Distr Table: ○ Ex. Freq of children’s height w class intervals – Use Grouped Freq Distr Table ○ Using Grouped Freq Distr Table - dots @ midpts ○ Could be made for puzzle scores from earlier  Usually label x axis at the midpoint of each interval • Simple & quick representation • Ex. for class scores - don’t need  Ex. For previous puzzle scores: No 0 to label each bar freq in end cats so would hit 0 in new end cats ○ To make freq chart from polygon:  Start w the dot w the highest x value ○ Usually data written in histogram forms • Nom or ord scales: ○ Score cats &X values not continuous ○ Use Bar Graph: Like histogram but spaces/gaps btwn bars  Indicate the scale is made up of distinct cats ○  Can’t assume all cats r same size ○ Ex. Number of cars ppl have – discrete not continuous, can’t have 2.5 cars so X can’t be 2.5 Shapes of Freq Distrs, Symmetrical vs. Skewed • Many pops are so large we can’t know exact # of indivs (freq) for any specific cat: Percentiles + [Percentile] Ranks + ○ So pop distr can be shown
More Less

Related notes for PSY201H1

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit