WEBVTT Kind: captions; language: en-us NOTE Treffsikkerhet: 91% (HØY) 00:00:00.000 --> 00:00:06.950 In this video we will look at graphical displays of two variables. NOTE Treffsikkerhet: 91% (HØY) 00:00:06.950 --> 00:00:11.000 Let us first open the data set. NOTE Treffsikkerhet: 90% (HØY) 00:00:12.800 --> 00:00:21.800 The kind of display that we can get obviously depends on the type of variable. If it's a numerical 00:00:21.800 --> 00:00:28.800 variable or categorical variable. If its quantitative or qualitative, in other words. NOTE Treffsikkerhet: 90% (HØY) 00:00:29.300 --> 00:00:44.300 Let us begin with two quantitative variables. Analyses, exploration, scatter plot. If you don't have 00:00:44.300 --> 00:00:51.600 this option on your Jamovi it means you have forgotten to install this scattermodule from the Jamovi 00:00:51.600 --> 00:00:57.900 library. Go back to the previous video concerning Jamovi installation and follow the 00:00:57.900 --> 00:01:00.150 instructions to install scatter. NOTE Treffsikkerhet: 88% (HØY) 00:01:00.150 --> 00:01:06.400 If you have scattered installed, this option will appear and so we can click on scatter plot. NOTE Treffsikkerhet: 90% (HØY) 00:01:06.400 --> 00:01:15.000 And scatter plot will ask us which two variables we want to plot against each other. So we can 00:01:15.000 --> 00:01:17.800 select, for example, NOTE Treffsikkerhet: 91% (HØY) 00:01:18.600 --> 00:01:27.700 matrices on the horizontal axis and letter knowledge on the vertical axis. NOTE Treffsikkerhet: 89% (HØY) 00:01:30.800 --> 00:01:40.800 This graph here is called the scatter plot and for each child, it displays one point and this one 00:01:40.800 --> 00:01:42.150 point NOTE Treffsikkerhet: 88% (HØY) 00:01:42.150 --> 00:01:50.800 is a combination of one coordinate for one variable and one coordinate for another variable. NOTE Treffsikkerhet: 87% (HØY) 00:01:51.300 --> 00:01:59.900 So this data point here, corresponds to a child, whose score on the matrices test at kindergarten 00:01:59.900 --> 00:02:01.700 was 15. NOTE Treffsikkerhet: 88% (HØY) 00:02:01.700 --> 00:02:09.900 And whose score in the letter knowledge test, the number of letters that this child new was maybe 00:02:09.900 --> 00:02:11.000 two. NOTE Treffsikkerhet: 83% (HØY) 00:02:11.000 --> 00:02:13.300 or NOTE Treffsikkerhet: 91% (HØY) 00:02:14.100 --> 00:02:18.700 However, whatever is around here. NOTE Treffsikkerhet: 91% (HØY) 00:02:18.700 --> 00:02:26.650 This data point corresponds to a child whose score on the matrices test was 10. NOTE Treffsikkerhet: 69% (MEDIUM) 00:02:26.650 --> 00:02:36.900 And who happened to know 10 letters of the alphabet. So each point is one child. And by looking at 00:02:36.900 --> 00:02:45.700 this graph we can see at a glance if the two variables seem to be related. If higher or lower 00:02:45.700 --> 00:02:53.000 values, in one of the variables are associated with higher or lower value variables in the other and 00:02:53.000 --> 00:02:56.850 what shape this relationship may have. NOTE Treffsikkerhet: 90% (HØY) 00:02:56.850 --> 00:03:03.100 So this is what you can do with two numeric, two quantitative variables. NOTE Treffsikkerhet: 68% (MEDIUM) 00:03:04.100 --> 00:03:15.400 What about two qualitative. Two categorical variables. Actually, the Jamovi doesn't have an option 00:03:15.400 --> 00:03:19.100 to plot a graph for that. NOTE Treffsikkerhet: 89% (HØY) 00:03:20.200 --> 00:03:29.900 But we can circumvent that if we really want to get a graphical display by using the RJ editor. NOTE Treffsikkerhet: 91% (HØY) 00:03:33.800 --> 00:03:36.450 and, NOTE Treffsikkerhet: 84% (HØY) 00:03:36.450 --> 00:03:49.300 designate which variables we are interested in. So notice I did not delete this hash sign here. If 00:03:49.300 --> 00:03:56.300 you deleted, type it again, and then open parentheses and type the two variables that you want to 00:03:56.300 --> 00:04:02.100 display graphically. So let's say sex and home language. NOTE Treffsikkerhet: 83% (HØY) 00:04:02.100 --> 00:04:09.100 You have to type the exact name of the variable, as it appears on the data sheet. If you make any 00:04:09.100 --> 00:04:14.900 mistake, then Jamovi will not understand it. It will produce an error message. NOTE Treffsikkerhet: 91% (HØY) 00:04:17.800 --> 00:04:21.800 And then, on the next line. NOTE Treffsikkerhet: 91% (HØY) 00:04:22.000 --> 00:04:27.600 You type mosaicplot. NOTE Treffsikkerhet: 78% (HØY) 00:04:28.400 --> 00:04:31.400 Open parenthesis. NOTE Treffsikkerhet: 91% (HØY) 00:04:31.400 --> 00:04:33.800 Table. NOTE Treffsikkerhet: 71% (MEDIUM) 00:04:33.900 --> 00:04:38.400 Open parentheses. Data. NOTE Treffsikkerhet: 91% (HØY) 00:04:38.600 --> 00:04:43.950 And then you have to remember to close both parentheses. NOTE Treffsikkerhet: 71% (MEDIUM) 00:04:43.950 --> 00:04:51.000 So now there's only one closing parentheses here which corresponds to this one and this one is left 00:04:51.000 --> 00:04:59.000 hanging. So if I click on the yellow triangle here, I will get an error message because Jamovi 00:04:59.000 --> 00:05:05.200 can not make sense of what I wrote. I need to close this parentheses. So now, everything is complete 00:05:05.200 --> 00:05:12.700 and I can click again on the run and I can get what is called a mosaic plot. This is not used very 00:05:12.700 --> 00:05:14.350 often and you will probably NOTE Treffsikkerhet: 91% (HØY) 00:05:14.350 --> 00:05:22.400 Never be asked to produce it. It can be useful if you want to compare proportions among different 00:05:22.400 --> 00:05:34.000 categorical variables. So here you can see at a glance that girls and boys have different 00:05:34.000 --> 00:05:39.500 proportions of majority and minority language. NOTE Treffsikkerhet: 91% (HØY) 00:05:39.500 --> 00:05:43.700 children. In particular NOTE Treffsikkerhet: 89% (HØY) 00:05:43.700 --> 00:05:52.850 almost all girls come from majority language homes, but there is a substantial proportion of boys 00:05:52.850 --> 00:05:59.850 from minority language homes. This is what this shows us in a graphical way. NOTE Treffsikkerhet: 91% (HØY) 00:05:59.850 --> 00:06:09.500 Okay, let's close this panel. And now go on to the situation for one categorical or qualitative and 00:06:09.500 --> 00:06:19.800 one numeric or quantitative, variable. Analyses, exploration, descriptives. NOTE Treffsikkerhet: 84% (HØY) 00:06:21.900 --> 00:06:29.900 Let us get rid of this descriptive stable now because you're not going to need it. NOTE Treffsikkerhet: 86% (HØY) 00:06:29.900 --> 00:06:33.049 And let's go to plots. NOTE Treffsikkerhet: 91% (HØY) 00:06:33.049 --> 00:06:41.250 So, one way to display a quantitative against a qualitative variable, NOTE Treffsikkerhet: 91% (HØY) 00:06:41.250 --> 00:06:48.400 is by using the quantitative variable here NOTE Treffsikkerhet: 76% (HØY) 00:06:48.400 --> 00:07:01.500 and the qualitative, the category, as a grouping variable. So we will look at scores on the matrices 00:07:01.500 --> 00:07:12.400 test split by sex and boxplot is the most appropriate graph for this kind of comparison. So here you 00:07:12.400 --> 00:07:18.700 can see at a glance if there appears to be any difference between girls NOTE Treffsikkerhet: 91% (HØY) 00:07:18.700 --> 00:07:22.150 and boys on this test. NOTE Treffsikkerhet: 91% (HØY) 00:07:22.150 --> 00:07:25.900 And you can see that there is great overlap NOTE Treffsikkerhet: 80% (HØY) 00:07:25.900 --> 00:07:36.150 between the two Sexes. The minimum and the maximum are about the same. The median is very similar. 00:07:36.150 --> 00:07:43.900 The boys of this sample seemed to be only slightly higher than the girls in the sample. But overall, 00:07:43.900 --> 00:07:50.050 there is great overlap, and there does not seem to be a substantial difference between these two 00:07:50.050 --> 00:07:51.700 groups. NOTE Treffsikkerhet: 90% (HØY) 00:07:53.600 --> 00:08:03.400 We can do other comparisons to. Let me go back to exploration, descriptives, and see if for example 00:08:03.400 --> 00:08:10.650 the Norwegian vocabulary of children differs by home language. NOTE Treffsikkerhet: 63% (MEDIUM) 00:08:10.650 --> 00:08:14.800 And asked for a box plot. NOTE Treffsikkerhet: 91% (HØY) 00:08:21.500 --> 00:08:25.299 And here we see that NOTE Treffsikkerhet: 91% (HØY) 00:08:25.299 --> 00:08:31.900 the central half of minority language children NOTE Treffsikkerhet: 89% (HØY) 00:08:31.900 --> 00:08:40.299 has lower vocabulary than the central have, and of course all above, the majority language 00:08:40.299 --> 00:08:48.000 children. It looks like with the exception of one child from this group that children from minority 00:08:48.000 --> 00:08:56.750 language homes have lower Norwegian vocabulary at kindergarten age. Which actually does make sense. NOTE Treffsikkerhet: 81% (HØY) 00:08:56.750 --> 00:09:04.300 And this is immediately visible in these kind of graph, where you can see all the quartiles. The 00:09:04.300 --> 00:09:12.300 minimum first quartile. Median. Third quartile, and maximum as an outlier here. The maximum is not 00:09:12.300 --> 00:09:14.200 an outlier. NOTE Treffsikkerhet: 91% (HØY) 00:09:15.300 --> 00:09:23.300 So these are the basic graphs you can use to display pairs of variables so you can see at a glance 00:09:23.300 --> 00:09:27.900 if there appears to be a relationship between the two variables or not.