nm5370: what we're going to do today is to take a hospital file a more up to date hospital file and we are going to import that into S-P-S-S so all i want you to then is boot up S-P-S-S can you find S-P-S-S alright down to the corner to start up to S-P-S-S and boot S-P-S-S quite up is that alright and so you get the screen same as oh what you're one like Farah i've forgotten your name Claire has got sf5371: i'm insulted nm5370: insulted are you insulted i see ok nm5370: what's up Ben Joe nm5370: oh dear we don't want you to do that thank you so if you just just hang on a minute and then we sm5377: nm5370: no we don't want you to do that thankyou su: he's only trying to be a star nm5370: lets go to file alright what i want you to d-, so what we are going to do sm5377: that's not clear nm5370: what's not clear you are the machine i think let me tell what we are going to do first of all and then we can see how we going to do it what i have prepared for you is a small dataset er i've stored it on the disc for you and what we're going to do is we're going to import that dataset into S-P-S-S and then in the second lecture this afternoon we are actually going to do some statistics because what you have this weekend this week so far is lot about policy a lot about er importing data into computers but there's no point in putting the data into computer unless we are going to do something with it that's fair isn't it is that okay so what we are going to do then is to get this dataset in and if i show you how to get the first one in then what we can do is you can do the rest now what i have done is to use one of the standard forms that i use back at work is that alright the thing about this particular form is if you look at it not only does it give you the variable but also gives you where the variable is in the dataset is that okay so if you look at age for example you'll find that in columns one and three one to three and also you'll find the data type because in S-P-S-S like S-P-, er like what you've got to do is to specify the type of data that you are going to feed in so it can be stored in the appropriate manner is that fair okay so what we need to do then i now stand round near Ben Joe for a bit because this his machine is is the nearest you alright Ben J-, what i want you to do now is to go up to new close that one and just just just pau-, just go right to very left go go to go to file file open it then when we come down it says read A-S-C-I-I data can you see A-S-C-I-I down there pull it down the fourth line down A-S-C-I-I data press that one and go across across to your right no A-S-C-I-I data and you find fixed columns can you see that press that one now what it will do it will have brought a table up can you see that table in front of you that's now inviting you to specify which variables we are going to feed into into the computer but before we do that we have to tell it which file we've got so i think we need to tell it a file does does it tell me where whe-, yes go to file hit the browse button and we need to go now to the P disk okay so lets go to let S-P-S-S open the P disk and down to the P disk open the P disk and you find a sub a sub file called hipe can you see that open hi-, dot hipe sorry it should be hes is that the wrong one lets try one of the other disks then Q disk oh dear oh disk oh dear and disk no not there yet oh disk its not there sm5377: nm5370: ha-, er i m looking for a file that says hes data see if you can find it i am having great problems at the moment sm5377: his data nm5370: hes data H-E-S data nm5370: having problems aren't we trouble is these are old Mac oh i tell you wh- , where it probably is well it should be on P drive alright it should be on the P drive it should be there sm5377: H-S-A nm5370: sorry sm5377: is it H-S-A nm5370: i can't see it wait a minute sm5377: oh i've seen it [laugh] H-S-A nm5370: look i see Manro there but can't see the other one i wonder if it's the wrong drive that's a big problem we're going to be can i get hold of the disk i want to pop upstairs for one minute sm5377: ok nm5370: alright then we'll see if we can g-, we'll see if we can put it into one of the other machine's is that alright we have a problem i won't a minute nm5370: now oh dear i'll logout login so i see C-D backslash i wonder do you have to logoff and logon again sorry nm5370: E-X-T is it what it is is it that's what we couldn't find okay so wait a minute sh-, sh-, so i'll just let me just copy copy er star dot E-X-T wait a mintute i can't do this way lets make a copy of it H-E-S ninety-eight dot E-X-T to H-E-S ninety-eight dot dat D-I-R can you see the dat file now ss: yeah yes nm5370: we're alright now are we okay so we go back now to ten minutes ago is that alright what i want you to d-, now do is to go to S-P-S-S go to read A-S-C- I-I data and pull up the panel are you alright there Ben nm5370: okay that's the one one up Ben no one up one up we you've gone to Q Ben that's the one we want go to H-E-S data there we've now got H-E-S ninety-eight so we can click that so we're okay are you alright so if you've now all picked off the right volume the right file what we can now start to do is to specify how to read in is that alright so what we need to do is if you have a look at the the th-, er screen above you what you'll find is its got a rather large box that looks like this and a number of small boxes that look like that the first field then is the first field which is on the which is on your handout here so what we are going to do her is we're going to put age in here and the maximum size of the word that you put in can only be eight characters long you can't make it longer than eight characters alright the next one it tells you in which record its going to be now what you can do with S-P-S-S is that you can actually take a series of single records like that each one of these refers to the same that that is the first observation the second observation and the third observation and what you can do you actually can get two pieces of data for the first observation two pieces for the second observation in which case that would then you'd need to have two records per observation for the particular use that we're going to make of dataset today we have one record per observation so this one here will always be a one we do then define where we're going to find the start of it which is in col-, column one where we are going to see the end of it which is column three so can you actually put those figures in there if you can sm5377: put age nm5370: yes er or you know or for a change you could write it in Nigerian age how's that then okay then the start column is in column one Ben sm5377: nm5370: that right just always one always one always one the next one is column one column one in there one down below that's column three and then below you'll then find you've got a window th-, three put three in okay right below that you'll find a window which says numeric as is can you see that sm5377: yes nm5370: now that's the type of data that you actually store in the record in this particular case its going to be numeric so that's fine we'll accept numeric as is is that okay okay so what we now do is that we now say er add can you see the add about about two thirds of the way down the middle just say add and what you'll find is then it says variable number one on on on record number one from column one to column three we're going to have age is that alright ss: yes nm5370: right let's do the second one then then you can have a go at doing the rest yourself what we'll do is this is now a date and dates are bit of devil to feed in really because the date is in three pieces if you look at it its in day day space month month space century century year year and S-P-S-S doesn't actually collect dates like this so what we're going to have to do is to feed in these three pieces of information is that alright so then this is going to be days only so the next one then is going to be day on waiting list day-w-l days of waiting list and the columns for days on waiting list will be seven and eight you get the idea for days that are in column seven and eight there's slash in column nine the month is in column ten and eleven a slash in column twelve and years in columns thirteen fourteen fifteen and sixteen do you get the idea namex you alright sf5373: right right nm5370: great what's up with namex sf5372: can you say that all again nm5370: shall i say it all over again okay le-, le-, let, well have a look at the board then and you can see what we're trying to do sm5377: they are all waiting lists nm5370: hang on a minute Ben su: could you move it a bit that way nm5370: okay we'll move slightly over here into the doorway how about that is that okay now the big problem we're faced with is there's a variety of ways in which dates are stored and in which dates are captured the tr-, the traditional one is something like day day month month year year or there's day day month month century century year year so for example that will be something like the tenth of december nineteen-ninety-eight for example and then there's things like day day slash month month slash year year or day day slash month month slash century century year year now we've got this form of date and what we know from our data our spec of our data that the date starts in column seven there and ends up in column sixteen there so in column seven and eight we have day and day and then we have we have a slash in column nine in column ten and eleven we have a month and month and a slash in column twelve and then we have century century year year in thirteen fourteen fifteen and sixteen so what we're going to have to do is to feed that date in three parts because S-P-S-S can't accept a date like that is that alright so what we do then is we c-, we first of all we feed is waiting waiting day day waiting list or waiting day list or something like that you can make up your own name waiting list month and waiting list year so waiting list day will be in column seven and eight waiting list month will be in column ten and eleven waiting list year will be in column thirteen to sixteen is that alright nm5370: so you can actually put those three variables in yes Ben sm5377: for this little box is year nm5370: yes you need to put it three times and add each one as you go you alright namex can you see nm5370: is it bit low it it is a bit low isn't it i'll move it up a bit nm5370: sorry sm5377: record one one is always one nm5370: one is always one and all alone sm5377: the secod one the column is seven eight nm5370: yes the second the next one down has got a seven in it and the bottom one has got an eight in it is that alright Ben how are you getting on Shauna sf5372: fine nm5370: Kerry Farah you alright sf5375: i'm fine nm5370: okay great how you getting Ben Richard okay great Farah good in order to get this in fairly quickly if we can if we get couple of the cari-, a couple of different variables in then i'll leave you to your own devices you see is that okay so y-, you'll end up with four rows in that big box on the right then and you'll find that with four rows in there sm5377: is it do we still put number of the first ones in there what have nm5370: i am sorry Ben that's great exactly right so that so you put a name here that will be that'll be month put in month sm5377: what i make here do i put something there nm5370: put month in M-O-N-T-H nm5370: and then start in column what sm5377: okay nm5370: starts in column ten sm5377: okay that's ten and eleven nm5370: ten in that one eleven on the one underneath Ben alright and again we're talking about numeric as is just stay onto that one and then you do same with the year put in the year that okay sm5377: yeah nm5370: you don't have to put these in any specific order so i am going to jump about a bit so that we can look at the different data types is that alright so we have done the first two then we're going to have a go at district next which is which is field number three now when we feed district in you'll find we are going to make a change again are you alright there Ben sm5377: yes nm5370: okay so with district wha-, in the big box on top then we'll put in district or D-H-A okay the box here will have one in it the box here will have the district starts in column nineteen goes on to column twenty-one but then stop because then we have a problem what's the problem any idea ss: no no nm5370: it's a character field that's right Pauline when you go down to numeric as is in S-P-S-S it causes a string field can you find the string field come down find the string field so it'll be a type string nm5370: have you found that so just highlight string and then say add and what you'll find is that when it says something like D-H-A like D-H-A one nineteen to twenty-one it will put an A in the end can you see that okay and that means S-P-S-S then knows that its character field is that okay alright Ben sm5377: yeah yeah nm5370: which is that su: it didn't put it as a string nm5370: okay what you do is you like that just to highlight it you say now remove it so put it back in again sm5377: yes nm5370: is that alright so its D-H-A the district health authority okay yeas that's always going to be a one put a one in there put a one put in what's what's what was in there then sm5377: nm5370: nineteen and twenty-one in here no you got too far actually Ben you beat me to it come come down to the next one now its a bit difficult one driving the keyboard and one driving the mouse i find but carry on twenty-one [laughter] it it's a combined effort here i can tell you [laughter] is that okay so what you find is the A behind it is that alright okay okay so i think what you can now do between you can you can actually bash on now put all the rest in now is that alright you found all the funnies okay nm5370: exactly right exactly right we'll have to put in three pieces then what we are going to do is to build the date up and i am going to show you how to do it in in S-P-S-S as well is that okay S-P-S-S is a bit funny about dates it really is nm5370: you okay Ben sm5377: not quite nm5370: okay well just bash on let me know when you got problem sm5377: no after the year what we do here is how do we feed the figure date in nm5370: have you got what sm5377: i thought you say we should go and put this in what are we supposed to do next nm5370: carry on putting all that lot in you've pu-, you've put in age you've put in date and waiting list you've put in the district now you put the end date now sm5377: okay nm5370: so put something like end something like E-N-D E day you've got to put that in and that's in start sm5377: column twenty-five nm5370: that's in start column twenty-five to twenty-six sm5377: yes nm5370: and that's going to be integer again remember remember to carefully change the type of data that you actually feeded in so that's add now do it again for month and day alright sm5377: yeah nm5370: okay Ben sm5377: okay nm5370: and how are you Richard okay sm5378: nm5370: why what's the problem sm5378: i've got the end day the end day starts with twenty-five nm5370: twenty-five yeah sm5378: so the so the the the day is the first two twenty-five to twenty-six nm5370: yeah then put a slash which is twenty-seven then you got the month month is twenty-eight and twenty-nine a slash is thirty then the year is thirty- one two three four i purposely built this dataset this way for you to make because it's one of the problems of trying to get data into S-P-S-S you know when it's in this has actually come straight out of a H-E-S machine you know a machine that's got H-E-S and i've not done any cleaning at all i've just taken the data that's all or some of the data for you to play with alright nm5370: that's good that's great that's er nm5370: the local specialty is the one which is one of the specialties which is a three digit code you're familiar with these codes in H-E-S aren't you well most of you you alright Farah sf5375: yes i am nm5370: i made a determined effort when i got here this morning i said her name is Farah [laughter] sf5375: you don't have to make an effort is my name that bad that you have to make an effort nm5370: it was a determined effort this morning nm5370: Shauna you okay there are you trying to work it all out Shauna are you sf5372: yeah nm5370: good for you that's great its little bit of mind-boggling at times but it's er su: well you can work it out as well su: i can't work it out on my own nm5370: no its good idea to write it down that's what i do i tend to what i tend to do is to do something like day day slash month month slash year year year year or centuries or som-, and i put in twenty-six twenty-seven twenty- eight twenty-nine thirty-one thir-, i tend to do it like that you know nm5370: and that's the way to space it out is that okay nm5370: now the beauty about this particular method is you can actually just pick out the figures you wanted you don't have to put all the intermediate ones as well nm5370: has it blown up what's the problem you can only put eight characters in its a maximum of eight characters is that alright nm5370: you've gone something like that is its a bit tedious putting this in but w-, w-, we need to put it in before we can do any before we can go any further you see it's good for your soul i i i tap like this i go twenty-nine thirty thirty-one thirty-two and i work it out like that you know it's surprising how people use their fingers it's incredible nm5370: exactly right su: these numbers nm5370: which ones sm5377: well this divided the range on all this nm5370: well its just to give some idea of what's happening you see because what i did on gender for example item number nine i did say there's code in there's one for male and two for female nm5370: is that alright so seventy-seventy is gender is it you've done quite well actually there er Ben sm5377: nm5370: gender or age are quite useful to get frequency you can actually you know recognize what the results look like you know something like diagnosis you are never sure what i did was i took a ten percent sample of two months data we had two months data and i took every tenth record down so that's what you've got there so there should be a reasonable distribution of diagnosis in that dataset su: oh god nm5370: oh no he's not going to help [laughter] nm5370: what's the problem su: no nothing nm5370: well you can actually remove it and put it back as string character is that alright su: i got changed nm5370: you're change worked did it that's fine su: i forgot we had another page nm5370: er see you forgot su: i was thinking i'd finished nm5370: you forgot the three extra items on the second page su: do you have to put an end of record marker on nm5370: i always put an end of record marker on the reason being is that you if you've got a variable length record sometimes its very difficult to spot wh-, so i always put end of record marker on i always do mm nm5370: not only that but you know exactly how many records you've got because its always on every record isn't it nm5370: sorry nm5370: well i'd put a Z in there actually so it'd be data type alpha er it just means that its much easier when somebody else views the dataset on screen and they can actually see where the end of the record is because if there's nothing in the end you use a lot of white space and its very difficult to find what is there you know if you are going to sort it sorts it faster and things like this with a m-, marker on it sm5378: nm5370: yes sm5378: nm5370: well there'll be a lot of white space in there as you say Richard that's right sm5378: just tap okay then nm5370: yes tap okay and lets see what happens now if you make a problem then what you can do is you can go back and read A-S-C-I-I data put this back up again and you can make some changes nm5370: thats what i wanted you to find out have you actually su: what do you call it nm5370: what's up Ben su: one month not one year nm5370: i see you going to you're going to er Kerry did you did you have an invalid field in any of yours sf5374: yeah nm5370: so what did you do did you change it to alpha sf5374: yeah i'm done su: it might be some old N-H-S number nm5370: there might very well be some old N-H-S numbers in there nm5370: okay what you do is you go back up to file go back up to er er no go back up to come out of there a minute open up the one that looks like a little funny television set which takes you back to the back up to the thing again nm5370: it looks like the thing with lots of little stingers on it you see i can see it Richard i can show you which one is it that one that's it now if you go over to file read A-S-C-I-I data again come back down to there and you've opened it up again so you can make the changes is that alright that's the way you do that yes Ben nm5370: what's the problem sm5377: teach me how to work these ones nm5370: sorry sm5377: er ho-, how to feed these ones how you work these ones nm5370: now what i did here was the first column here will be column seven that's column eight that's column nine that's column ten that's column eleven that's column twelve that's thirteen fourteen fifteen sixteen so it goes from seven to sixteen but its divided up like that can you see sm5377: okay then what of the rest of this nm5370: th-, th-, th-, th-, others will be alright lets see how far you've got its okay lets see how far you've got lets see what its done nm5370: well certainly somethings it doesn't like nm5370: you've got a lot errors there Ben can you see sm5377: look at errors yes nm5370: er so what we've got to do is go back to there and then go back to there down to read A-S-C-I-I data and then fix columns and have a look at it again so that i know where errors are sm5377: so to fix column nm5370: i am sorry Ben sm5377: what do i have to do to check nm5370: to change things what you do is you actually highlight a corner like like that then you either remove it or can you ch-, change it just say re-, remove it and put it back in again sm5377: okay nm5370: okay sm5377: this okay nm5370: er alright how are you getting on Shauna sf5372: fine nm5370: how are you getting on you two alright sf5374: i've still got errors even when i change the er nm5370: what's the errors then what's turning up then Kerry sf5374: i can't really tell which one it is nm5370: you can't tell because normally it tells you to start number because from that you can work out which field it is causing the problem sf5374: i think its one of the specialty fields actually nm5370: one of the specialty fields is it okay sm5377: you blow up this one can't do it a thing anymore nm5370: one way to quieten the machine down of course is to make everything alpha field and then it doesn't complain how er Richard you alright sm5378: it asks me if i want to save contents of data su: i need to give it a filename nm5370: er give it a filename nm5370: i thought you already changed the filename we've already done it once su: windows you create a file er you use it this one you save it then give it a name afterwards nm5370: okay i see we don't want to save it anyway whatsoever all we want to do is to correct it that's all we're doing is that alright its still a puzzlement is it nm5370: it shouldn't need to just say close it doesn't need to do that because you're not trying to save it you see nm5370: er right Shauna are you ok sf5372: mm mm nm5370: Marion and Pau-, Pauline you alright su: we're just trying to work out the er nm5370: okay that's great fine that's good su: one two three four five nm5370: how are you getting along Stella su: it wouldn't let me enter zero nm5370: what do you mean it wouldn't let you enter a zero nm5370: that's always always one su: i see i forgot nm5370: is that alright nm5370: i am trying to actually catch this myself so i can put a bit of data in so we can actually do something sm5377: sm5377: mistake correct it nm5370: you say remove it and then put it back in again Ben is that alright highlight the line with the mouse sm5377: to remove it nm5370: okay alright let me do it for you then you can do it yourself sm5377: line two oh it's remov-, it's removed nm5370: its removed it sm5377: yes it was you see it's gone so what do i do to get it nm5370: well you' ve got to to put it back back in again so you go to there you start to type it again is that alright su: i can't get my table i can't seem to click on o-, okay nm5370: what do you mean you can't seem to get on okay su: its not coming up as okay i can't i can't nm5370: you can't delete it to add that one just to add it first su: nm5370: which is that su: oh yeah nm5370: it should should be okay now nm5370: come back to where where was the a-, where was the go back and put your put your P cursor on one of these one of these ones over here su: twenty-one and twenty-two nm5370: that okay that's not going coming up again oh well close the file make sure you've got a file to do it with alright just ty-, double click that su: oh yeah nm5370: it didn't have a file to apply it to you see that was the problem that okay su: yeah it has loads of problems nm5370: beset with problems aren't they su: yeah su: millions of them nm5370: what was that su: postcode nm5370: postcode is alpha now why should there be a problem with postcode if its alpha there shouldn't be a problem with it su: i don't know nm5370: mm what's the problem su: i am not quite sure obviously su: i known neither am i nm5370: i think you've set it up to numeric your postpode go back and have a look go back to here over to file down to read A-S-C-I-I data again clo-, go go go pick mm yeah go back to here go back down to postcode have a look su: do you think that's why nm5370: where's your postcode nm5370: its easiest to look at the numbers there actually su: after gender nm5370: seventy-three nineteen sm5378: yeah nm5370: you're going to set up as an alpha there then that's the problem its try-, its trying to make a number out of it you see that's the problem but i don't know how many cases are on this particular file so what its doing is trying to set a number up su: we don't understand what this means nm5370: you don't understand what it means what wh-, what what means well what its trying to say is as its gone through your file converting it now for command line nine case three-hundred-twenty-eight it says it says its found a thing called thiong called three G in starting column forty-seven now what's starting column forty-seven that's local specialty it's found the specialty of three G so it looks as if specialty has now got to be coded with alpha rather than in integer is that alright you'll find you'll find a specialty like four- ten-twenty two-hundred-twenty five-o-five and so somebody's put a three G in the middle of it haven't they nm5370: yeah okay so what you need to do is go back and find that convert it to alpha by by redefining it and then have another go okay what you do then er ge-, get rid of this get oh you've found it have you sorry yes that's okay you can do it you can do it alright Shauna can't you yes Ben what's up sm5377: i've got to take this i want to fix it between nm5370: it does it automatically just say add it will do automatically sm5377: okay nm5370: does it automatically for you nm5370: i want it clean this dataset because w-, because when we've had cup of tea we are going to start to do some work on this you see so the idea was to get it as clean as we get it get all the things properly defined alright what's up Richard you look really puzzled you are puzzled are what's puzzlement can we share the puzzlement nm5370: alright sm5378: and when i go back in to redefine that i still get the same thing action saying do you want to save this file nm5370: is it okay yes er su: nm5370: well now what its doing is if you say you want to redefine it you don't want to save it that's what it is you just say no and when it says do you want to save the file you've got so far it says and the answer is no i don't want to i'll recreate it as i want it that alright because thats what Shauna's had a few minutes ago the answer's no is that ok yes er su: i don't kn-, i don't know what any of this means nm5370: well now what its telling me now its its telling you that M here in current split file its found that thing in what starts in column seventy-three it loooks as though you still haven't defined your er su: postcode nm5370: postcode properly it looks as if its still having problems so lets kill this k-, kill that one up there where do we kill that you alright nm5370: go down to postcode where's postcode now yes it hasn't got an A on it at the end you see su: oh i thought i changed it nm5370: okay let's say change su: yeah nm5370: is it is it alright now i say okay the answer's no you you don't want to save it answer no lets see what happens now is that better su: mm mm nm5370: i've just done it su: i have done it nm5370: you sure it is in there what's the error now su: er well it still says er in seventy-three but i think that data's all nm5370: how many arrows are there su: it still says its in column seventy-three er er the thing that isn't coming up nm5370: it could be that it could be that that's the old stuff you see you've got the new stuff carry on going down su: forty-seven nm5370: carry on going down su: that's it forty-seven mm mm nm5370: forty-seven now what's forty-seven its local specialty i think you'll have to make that an A as well is that okay su: okay nm5370: sorry nm5370: you got three G well it probably it wouldn't have put in you see if it couldn't have got it it would have complained bitterly that it has found one for three G and when it finds one it can't do anything with it cause it doesn't know what to do with it so it hasn't put it through it's put a dot in it instead so thing is to go all the way back to redefine local specialty it's the alpha field rather than a numeric one su: wait wait wait nm5370: how do you do it okay go back nm5370: go to file file down to read A-S-C-I-I data tap A-S-C-I-I data down to that one now come back and find it on there you make an alpha field instead is that alright su: the first one's wrong sm5377: i finished correcting nm5370: you finished correcting Ben nm5370: Pauline are you alright is it all in its all in is it how about you Marion sf5376: yeah nm5370: I know Kerry's in Farah is in how about Richard okay so the backline the back row is all there is it that's great sm5377: i i finished corrections my printer okay now i don't where to go now nm5370: the answer's no you don't want to save it su: can you help me sm5377: no but i want to save that's all i really need nm5370: you say no okay now it's going to reload it for you then there you are well that's great that's really empty this time sm5377: no i wanted this nm5370: you wanted what its still running still running though its processing is it not running for you nm5370: there read A-S-C-I-I data read A-S-C-I-I data fix come on A-S-C-I-I data fix columns and what you do now you just say okay yes if you find the variable that's okay you say okay try again nm5370: its executing now lets i see well you've got some data in there Ben sm5377: but that's that's not alright that old one nm5370: that's an old one is it sm5377: old old one nm5370: okay well lets go back and scroll back i think what the big problem you've got the some problems in here you see this is the big problem sm5377: nm5370: i spent all last evening when i got home doing this building this file age one two three okay that's fine date nm5370: yeah that's alright don't worry don't worry you've got some funny things in there month yeah redefine that again month year D-H-A district end day end date start you've got most of these in hereokay okay nm5370: okay okay its bumping all sort of things in there its got a window lets have a look at the viewer sm5377: er you see nm5370: what's that sm5377: that's working here nm5370: er the one eleven-o-two it says i'm not going to do anymore it says nm5370: it's okay you you're alright actually there's few missing you got a few missing i think leave it for a minute just leave it for a minute come on i want to give another backup you're alright so what we need to do then now is to look at have you st-, have you got any output from this particular package yet have you tried looking at the output okay first it got very high quality output what i wanted er to do then is to start to look at some very basic frequencies you've done it you've already you've done age you've done gender okay what we need to do is to start to look at the data because we are going to compute some new data is that okay because what we've done so far is got our dates in the form day day month month century century year year we've got to put that into a reasonable format is that okay so let me just boot my machine up and then we can see what we are going to do with it all i am going to put in my machine is just the one date that's all i am going to use in my machine all i've done is put together now a file which contains the three parts of the date and what we need to do is we need to build to compute a new date because the date as it is now is pretty useless because what we need to is we've got two fields in o-, on output on the dataset now one is called end date and the one is called start date er so what we can actually do is that we can if we got if we've got an end date and from end date we take start date then the result will be in length a length of stay in days well what we've got really is we've got our dates in three pieces haven't we is that okay so what we've got to do is to put those dates together and there are ways of doing this and what and what oh even i've got a problem now what problem have i got okay thank you yes i got it nm5370: as you just heard the dates are missing so i've got something like end dates end days end month and end year what we want to do now is to put that together into some sort of date so we can say the composite end composite end minus minus the start then that's going to equal then the stay is that alright so what we have to do is to work do er is to put those together what i want you to do then is to go to transform can you er at th-, the very top can you see on the very top column called transform can you see that and bring that down and in there you will see er yeah is that the one that i want yeah transform is the one i want nm5370: yes what i want you to do is to bring down compute so you will end up with a window called compute er so you come er you pull down transform and you pull down compute can you see that and you'll end up with a window a small window that looks a bit like this it's it it's a window like that but the window across the top here has got a sort of keyboard in here yeah and window down there on the on the right hand side alright can you see the one i've got have you got that one alright is that okay what we can then do is we can take certain variables and we can put them together in the left window here it actually gives you a list of all the variables you've got so far and all i've put into my machine are the four variables i put in the age the end days the end month and the end year if you gently come down the right the the the lower right hand window now if you gently come down there you'll find there's some functions with days in them okay date in them can you see the date functions can you see that alright now there's one of those called day month year or month day year or there's various ones of them but i think what we ought to is to try and put together a date called day month year what it will do is it will take the days and the months and the years and form a new composite date is that alright so what i want you to do then is to highlight this particular option called date D-M-Y nm5370: so its going to be date like this D-M date sorry date D-M date dot D-M- Y can you see that and what its looking for its looking for something like days months and years is that okay can you see that Marion or is it a bit too low i'll keep on writing at the top of the board so what i want you to do is to highlight that and what it will do is you've got to give a target variable now you've got to give it a target name what shall we call it shall we call it new end or something like that is that okay call it anything you want i am going to call mine new end for example alright and then then if you highlight the day dot M-D-Y and bring that up what you'll find it it'll actually push it up into the top right hand box can you see the little arrow that pushes it up into there can you see that su: oh there nm5370: yeah so what you'll find here then is in this little box over here i've got new end like that and in the big box over here i have now got a copy of date dot D-M-Y and open brackets its got some bits can you see it and in there you'll find its got three question marks in there and what you've done now is highlighted the first question mark can you see that so if we now go down to the little left hand box can you see the next to the first question mark we put er no i got it the wrong way round get down oh come out sorry i'll just do it again transform compute and you end up i pulled the wrong one down when i was talking er so what i want to do is to go for the date which is D-M-Y put you up there so in the first of those question marks you put in you go down to the little left hand box and highlight the the variable that's got the days in it first of all and you should be able to sub that particular variable up into there so it ends up with something like date D-M-Y and in there its got my days or something like that it puts that in and then you highlight the second of those little question marks now with y-, with your mouse and you do same with the month push that up and you do the same with the third of the question little question marks and you put the end th-, the year up as well so what i've ended up with now is date dot D-M-Y open brackets end days comma end month comma end year or whatever variables you named is that alright now if you now say okay you now say okay what it'll do is it will beaver away and calculate oh my-, mines all come up as one point one each of them plus ten what's it come up like su: nm5370: what's your come up with nm5370: what's your come up with su: mines got an error message its got incorrect variable made must be more than eight characters su: nm5370: you it Marion or are you alright well mine's got an error which is actually marvellous sf5376: yeah nm5370: what's going on here are you all busy or something you're all busy su: don't even ask nm5370: don't even ask alright i won't even ask nm5370: what what's that going on mm what what have you got i've got a really funny one on mine what have you got su: is that is that to do with because the year century century year year nm5370: may-, maybe that's what it is actually there's another date function down there which just uses day and month we could just try the day and month and see if that comes out alright what do you think shall we try that one instead it obviously doesn't like doesn't like day day space month month space century century year year does it but we can what we can do is that we can compute a new date which is dead easy if we were to take nineteen-hundred away from all days that would actually take us back to just year year wouldn't it su: it will nm5370: or we could do it with compute a new end year yes have a go at that compute a new end year you say new end year or N end year equals end year minus nineteen-hundred have a go at that i am not being very kind to you on this one i don't want to be kind because i want you to find your own way around this package you know what's up Richard you look very puzzled sm5378: i'm alright nm5370: if you want to clear that thing you've got to press reset all the time you'll find a reset in the middle of it you just highlighted and got rid of it did you okay then