kayden visualization is a critical piece

of modern-day statistical analysis their staples are helpful still you have

to eyeball to the output to figure out what the trends are the visual display

is much more easily understandable so using the same data sets of teaching

evaluations we ask this question to instructors teaching single credit

courses get higher evaluations and we see that yes they do be mean evaluation

when plotted as a chart you see that those who teaches single credit courses

the average teaching evaluation is slightly higher let’s look at some other

graphics before we attempt to repeat the same in SPSS in this graphic you see

that we have plotted the mean or average teaching evaluation score for tenured

professors and then we have added another gender dimension to it so green

color is males blue is females you would notice that untenured professors male

have higher teaching evaluations and for tenured professors the difference in

teaching evaluation is not that pronounced as it for as it is for

untenured let us repeat the analysis in SPSS we click on graphs chart builder

I’m just going to reset it so it’s clean bar track the bar chart template here

click on gender or female instructor on the x-axis teaching evaluations go drag

and drop it on count and click on ok and here’s the graph now we want to do the

same for tenure but adding TG adding gender as a additional dimension I say

graphs charter reset this time I pick the different template click on x-axis

we pick tenure and teaching evaluation and run count drag and drop and then

female on the clustering on X and then ok and there we go if one has two continuous variables such

as the beauty score and the teaching evaluation both are continuous not

categorical the best way to depict them is using a scatterplot so you’re looking

at a scatter plot where beauty school is mapped out and x-axis and evaluations

are laid out on y-axis and the color of the dots represents male and female so

blue is female Green is male another way of presenting the same data is to have

two side by side panels same data beauty and evaluation beauty on the

x-axis teaching evaluations close on the y-axis but we are plotting a separate

chart for females and another one for male instructors so if you were to

recall the other one we have all the data set in one chart differentiated by

color but if you think that the fat things are not very easily detectable

then you can plot them separately and lastly we can fit a regression line to

it this is a lowest line that we have done we will discuss this later but just

to show you how to get these results I’ll take you to SPSS to generate a

scatterplot i click on graph chart builder click on scatter and pick the

second so first let me reset and then second template on x axis I’m putting

beauty and normalize beauty school on x-axis and dragging the teaching

valuation score on Y axis and the color by gender so I’m taking the gender

variable in frontier click OK and this is the and same graph that you saw

earlier in order to do multiple panels we recall the same dialog box reset and

say scatter pick the appropriate template drop teaching evaluations score on Y

axis drag normalized beauty score on x axis so this is the same graph as before

but now we take groups and select the column panel variable and shows up here

and now we say female in instructor variable here so now you’ll have

multiple side by side panels click on ok and here you go in order to fit a

regression line we have to edit this graphic to edit it double click on this

graphic and it will open up an editing window I will select one of these

options at fit line two at total so I click this it opens up a dialog box I

select Louis lines and then I can draw these but I click on lines pick a color

of my liking so I say line red color and I want to make it a little thick so I

said to wait is to this type of lines apply close this here and then the lines

are select notice that it has 50 regression line and added the equation

to it which is not what I intended to i’m going to double click on it and

click on line and go back and say Louis now I fit pick the right one I don’t

want any labels on it so I simply select that apply and now it double click out

of the editing mode and now I’ve got these regression or the best fit line

attached to it here we are looking at the average teaching evaluations for

differentiated by gender male and female and further differentiated by tenure and

this is called the box plot we can have easily plotted just one box plot that we

have got four versions of it differentiated by gender & tenure we

will explain box plots later but let’s see how we can generate this graphic we

return to SPSS graph chart builder reset click on gallery pick box plot and you

learn multiple dimensions so we just click bring it here and then on x axis

we have gender so female instructor here and y-axis is the teaching evaluations

score and then we want the groups by rows and the panel is tenure you drag and

drop tenure on panel click OK a boxplot

differentiated by tenure and gender