Filter and PivotTables in Excel

Transcription

1 Filter and PivotTables in Excel FILTERING With filters in Excel you can quickly collapse your spreadsheet to find records meeting specific criteria. A lot of reporters use filter to cut their data down to size. Here are some examples: A reporter gets a spreadsheet with data covering her entire state. By filtering she can quickly pick out only the counties in her coverage area. When working with school test scores a reporter wants to look only at schools with enrollment greater than 500 but less than 1,000. He can do that by filtering using numbers and ranges. A political reporter can use filters with campaign contribution data to find all records that came from donors in a particular city or zip code. A crime reporter got a spreadsheet of incidents from the police department. She heard that there was an increase in crime on one particular street. She can find all of the incidents occurring on that street by filtering the address column to find any records containing that street name. To turn on the filters simply click anywhere in the data and then look under the Data tab for the Filter button. The icon looks like a funnel. Click on it and dropdown arrows should appear next to each column header. Figure 1 If you click on one of those dropdowns, you should find a unique list of all of the values contained in that column (which is a great way to get to know what each column contains). You can click on the box next to Select all to uncheck all items. Then select only what you want. The example below shows a filter to get only the records for the National League in a spreadsheet of baseball players. Once you re happy with your selections, close the box. 1

2 Figure 2 You ll know the filter is working by several things: 1) The dropdown next to the heading of the column you re filtering on will become a funnel. 2) All of the row numbers become blue. You now only see the records meeting your filter criteria. None of the other records have gone away, they are just hiding until you either change your filter or turn it off. 3) At the bottom left-hand corner of your screen you ll see a record count. This tells you how many records have been selected out of the total number of records. This can be very useful during an analysis: 2

3 If you want all of the records to come back you have two options: You can click back on the dropdown next to the heading of the column you are currently filtering and click on Select all. Or, you can just turn off the filters; often this is the easiest way to go because you can think of the filter like a light switch. You flip the switch and the filters appear, you flip it again and they turn off. To turn off the filters just click on Filter (the funnel icon) under the Data tab. It will clear all of your filters, bringing back every record. Now that we ve covered the basics of filtering, explore filters on your own to see all that they offer. Here are some highlights: Put filters on multiple columns at the same time. When working with number columns explore the drop-down number filters under Choose One. Options here include Greater Than, Between, Less Than and Top 10. Figure 3 Text columns also offer some useful filtering options. Find records that begin, end or contain certain characters. The Contains option is especially useful if you have to deal with information that is commonly misspelled or spelled inconsistently. For example, St. 3

4 Louis can be written many different ways, including St. Louis, St Louis and Saint Louis. None of these are wrong, but Excel will see them as different items. They all have Louis in common so you could use the Contains or Ends With options to grab all of the records at once. Figure 4 4

5 GROUPING We re journalists, so we re often concerned with answering questions like: Who is paying the most? Which county had the most? How many? What is the most common? When you start wanting to ask your data questions like this and a simple sort won t do, you re going to need to use pivot tables. One of the main concepts to grasp with pivot tables is grouping. Imagine if I had a traditional deck of cards, shuffled them up, took two cards out and put them in my pocket. You d know I had 50 cards remaining, but if I asked you which cards were missing, what would you do? You wouldn t be able to just look at the deck of cards and tell me that I had the Queen of Hearts and the 8 of Spades in my pocket. You d put the cards into piles, each containing one type of card, such as Kings, Aces, 7 s, etc. After you d gone through the 50 cards you d be able to see which pile had fewer than four cards to discover which ones are missing. Excel is going to do the same thing for us with a pivot table. Before learning this skill many journalists have simply used a piece of paper and a pencil to tally up things and report on them. Take, for example, this listing of individuals and firms barred from doing business with the World Bank (See Figure 5). One reporter wanted to write a story looking at the number of debarments for his country compared to others. He went through the list of more than 500 records of companies and individuals and kept a tally of the number of records for each country. The story isn t impossible without Excel but analyzing this same information in a spreadsheet can drastically cut down on the time and increase accuracy by doing the math for you. Follow the steps below to walk through a basic pivot table using WorldBank.xlsx. The data contain the name, address and country of the debarred individuals or firms as well as the ineligibility dates and grounds for debarment. 5

6 Figure 5 Figure 6 6

7 To find out the total debarments for each country you ll need to put the countries into groups using a pivot table. BUILDING A PIVOTTABLE First, highlight all your data: select A1 and hold down Shift + Command, then hit the right arrow (which should highlight all the headers) and the down arrow (which will highlight all the rows). Next, go to the Data tab and look all the way to the left. You should see PivotTable. Click the small down arrow and choose Create Manual PivotTable. Figure 7 The Create PivotTable window should open. It has two pieces to it: 1) You re asked to select the data you d like to analyze with your PivotTable. This is why you should select your data in advance. The Table/Range: information in Figure 8 is showing us exactly what we selected. It looks funny, but really it s just saying that we selected cells A1 all the way through F608 in the sheet called WorldBank found in this workbook. 2) Excel wants us to tell it where we d like to put the PivotTable. By default it selects New Worksheet. This is good because we don t want the PivotTable to just appear right on top of our data. 7

8 Figure 8 If you follow our steps you should always be able to simply click OK in this window, but it s still good to understand exactly what Excel is doing. After you click OK, Excel pops you into a new sheet with all of the tools you ll need to build your summary. There are two pieces, the various boxes on the left and the PivotTable Builder on the right. The boxes on the left are where your summary or chart will appear and change each time you do something in the task pane on the right. See Figure 10 for more information on this task pane. Figure 9 8

9 Figure 10: PivotTable Field List Task Pane We call them column headings, but Excel calls them fields. This is a list of all of the column headings, or fields, in the data. These boxes allow you to do some more complicated things in your PivotTable. We won t use them in the introduction to the concepts, but we ll explore them in some of the in-class and homework exercises. This is where you put your groups. Think of this as the labels for your chart. It s where you d put counties, team names, cities, etc. anything that will be the group. We typically start PivotTables in this area. Just drag a column heading from the field list here and the PivotTable box will automatically contain a list of unique values found in the column you chose. Think of this as the place where math happens. Anytime you want to find the sum, average or total number of records for a category you ll work in this area. Just like we ve done with other datasets, frame your analysis with a question. In this situation we want to know which country has the most firms and/or individuals on the debarred list. To answer that question you ll move Country from the Field name list to the Row Labels box. As 9

10 soon as you drop Country under Row Labels you should see a list of country names appear in the PivotTable box. This list is alphabetical and each country name should be listed only once. Figure 11 Next, we ll want to count the number of debarments for each country. Remember that each row in the spreadsheet represents one firm or individual debarred. To count up the totals by country, drag Country under the Values box. See Figure

11 Figure 12 The last step is to get the country with the most debarments on the top of the list. For this we ll need to sort. Sorting is different in PivotTables than sorting in a regular sheet. Here, all you need to do is click on any number next to a country, and use the dropdown tool on the sort icon to select Descending (Figure 13). You can also sort the records alphabetically by clicking on any one of the country names, selecting the sort icon, then whichever option you prefer. Figure 13 11

12 Sorting by the number of debarments brings Canada to the top of the list, followed by the United States, Indonesia and the United Kingdom. Notice that Excel creates a Grand Total row at the bottom. This total should equal the number of records in your original spreadsheet. Figure 14 12