in the previous article "5 reasons to tell you why JMP software is more suitable for you?" " , we briefly introduced the advantages of JMP software for non-professional statisticians: JMP and its friendly menu interface, powerful visualization tools, rich data cleaning functions, systematic statistical analysis methods, interactive results display methods, etc. , very friendly to non-statisticians.
In order to help clinicians learn how to use JMP to efficiently carry out data analysis and improve the efficiency of daily work and publication of papers, Dr. Feng Guoshuang, a senior JMP user and senior statistician, and his team will share a series of statistics and data analysis, JMP operations, etc. Practical content, a classic topic in each issue, to help you master a new skill. It is worth noting that these topics are not specific to clinicians, but apply to all those who use JMP software to conduct data analysis.
This article is the second installment in this series. Dr. Feng and his team introduced the first step of JMP data analysis: importing data. This article will take Excel as an example to teach you how to cleverly import existing data, which is very similar to the import method of other types of data sets (such as SPSS data, csv data, etc.).
Three ways to easily import data
Three common ways to import data into JMP are: copy and paste, directly open existing data, and import through the JMP picklist in Excel. Let's take a look at them one by one!
(1) copy and paste
First, open the JMP software, there is a small icon similar to a data set under the function table [Document], when the cursor moves to it, it will display [New Data Sheet], click this small icon (Figure 1), and you will see Go to the interface in Figure 2. At this time, go back to the Excel data set, find the data set you want to analyze, select all the data cells you want to analyze, and copy it.
Here is a little trick. If you have a lot of variables (rows) or observations (columns), it is a bit troublesome to select by dragging. You may wish to click the small triangle in the upper left corner of the table to select the entire page. Then go to JMP to copy. Don't worry, JMP can accurately identify cells with data content.
Figure 1 Create JMP empty data set
Figure 2 JMP Empty Dataset
After the data is selected and copied, return to the newly created JMP empty data table. Here, it is not a simple paste or ctrl+v, because when we copy in Excel, the first column is the variable name, if you paste it directly, the variable name becomes the first column observation in the JMP dataset Already, this is not feasible.
The correct way is to click [Paste with column name] under [Edit] in the menu column in JMP. At this time, the first row of your Excel will be placed in the position of the variable name in JMP. Of course, you can also use shortcut compound keys, but instead of ctrl+v, you can use ctrl+shift+v.
So far, the first method of importing Excel data is over. Is not it simple?
(2) Open the existing data directly
This method of importing Excel into JMP may be the most commonly used method in daily practice. First, still open the JMP software, select [File] and click [Open (O)...] in the drop-down menu, and then locate your Excel data set in the pop-up file browser (Figure 3, if you can’t see it For Excel type files, just select the Excel file in the drop-down menu on the right side of the file name. JMP can import most of the commonly used database files, including text files, PDF files, SAS program files, R codes, and MATLAB codes , HTML file, Shape file, JSON data file, etc.), click to select, and then open.
One thing to note here is that when the file type locates an Excel file, a column of small characters and an option will appear above the file name, prompting whether to force the first column of Excel to be used as a label. Generally, we will choose "best guess". That's right, JMP can always correctly judge whether the first column is a variable name. Of course, you can also choose "Always" or "Never", depending on your profile.
Figure 3 JMP opens the existing data set interface
After clicking "Open", you will enter the preview stage before the data is officially imported, as shown in Figure 4. The upper left corner is a preview of the data, where you can drag the scroll to see if the number of observations and variables is correct (as long as your Excel format is correct, JMP will not make mistakes).
There is another more practical function here in the preview window. You can select a row and click the right button to specify the input format of the data, such as changing the decimal place of a numerical variable, changing it to a percentage format or changing it to a date format, etc. Of course , these can be operated after the data is imported into a JMP data table, and if you make a mistake here and want to start over, the bottom "Restore Default Settings" can help you.
The choice of sheet on the upper right is the worksheet in Excel. If there are multiple sheets in your Excel sheet, please pay attention to confirm the name of the sheet here. If you want to import all the sheets in the table, you can choose to select all here, then as many sheets as there are in Excel, JMP will import as many data sheets.
The lower left is the setting of a single worksheet. The default setting is that the row title starts from the first column, the number of columns with titles is 1, and the data starts from the first row and the second column. If your data format is special, then here is Modify as appropriate. Others we maintain their default state just fine.
Everything is ready, now, just click [Import].
Figure 4 Import EXCEL data set interface
(3) Through the JMP option list in Excel (this is a unique function of Excel)
This method may be the easiest to operate of the three methods! Under normal circumstances, after we have installed the JMP software, the JMP plug-in program will also be installed in Excel. Open your Excel data sheet, and you can see the JMP option list, just like the EndNote plug-in program we see in Word (Figure 5).
Figure 5 JMP plug-in program in EXCEL
Before generating the JMP data table, we need to click Preferences to set it up. After clicking [Preferences], the interface in Figure 6 will pop up. You need to check "Use the first row as the column name", that is, use the first column of the Excel data as the row variable name, click OK, and then click [Data Table] next to Preferences (Figure 7), you can directly generate the JMP data table.
Figure 6 Settings of the JMP plug-in program in EXCEL
Figure 7 Click "Data Sheet" to automatically enter the JMP software from EXCEL
The above are the three most common and easiest ways to import Excel data into JMP. Data import seems simple, but if you master some tips, you can get twice the result with half the effort. Based on our many years of experience in using JMP, here are two considerations for importing data.
(1) The data in the original data must not have any non-numeric symbols
In JMP, the "continuous" variable name is displayed as a blue triangle, and the "nominal" variable name is displayed as a red column (Figure 8, the ordinal type is displayed in green, there is no ordinal variable in the figure, therefore not shown). After importing data, if the software recognizes that the variable is of numeric type, then you can change it to any type, and if the software recognizes that a variable is of character type, then the variable can only Type".
Figure 8 Schematic diagram of different variable types
Under normal circumstances, after the data is imported, JMP defaults all values as "continuous", and any non-value in any variable (such as accidentally writing 8.6 as 8.6, or a cell with "+", " #" and other non-digit symbols, etc.), are considered to be "nominal type". Therefore, if a variable should be a numerical value (such as age, height, etc.), but it is found to be "nominal" by default after importing, it means that there must be non-numeric letters or symbols in the variable (such as commas written as decimal points, etc.), then It is necessary to check the original data, otherwise the nominal variable cannot be analyzed as a numerical value.
For example, the BMI in Figure 9 is all numerical values in the original data (such as Excel), so it is no problem to directly preset it as a continuous variable after importing.
Figure 9 When the original data is a numerical value, it is defaulted as a continuous variable after importing
In the BMI in Figure 10, in the original data set (such as Excel), due to carelessness, one of the numerical values has a decimal point replaced with a period. At this time, the cell that looks like a numerical value is not a numerical value. So if you import it directly to JMP, you can import it successfully, but because there is a non-value (21.6) in BMI, the default is a nominal variable.
Figure 10 The existence of non-numeric symbols in numerical variables causes them to become nominal variables after importing
So be sure to pay attention, when you import into JMP from a data set such as Excel, if you find a variable that should be a value but is preset as a character variable (marked in red), there is only one possibility: it must be yours In the original data, at least one of the variables contains non-numeric characters, which may be inadvertent things such as periods and commas, or you are used to adding characters such as "-" and "/" to spaces, or it may be Some software defaults to fill the empty value with "#NULL" and so on. In short, anything that is not a numerical value will not work.
(2) JMP imports the most relaxed restrictions on variable names
At the end, I have to mention another huge advantage of JMP when importing Excel data - you don't have to worry about your variable names being unrecognizable by the software! At present, most statistical software has restrictions on variable names, such as no brackets, no special characters, and so on. For example, age (age), such a variable name cannot be directly imported in other analysis software or programming software, and all prompts that they do not know (unrecognized). This is especially troublesome when drawing a picture. Either the picture has to be edited, or the display method of the variable has to be changed through the label form, which is an extra procedure.
In JMP, there is no such concern. Whether your variable name is pure numbers, pure letters, special symbols or various strange combinations, JMP can import it as it is. Just change it to the format required by your own article. Although it seems like a very simple small improvement, it actually saves a lot of trouble for publishing the article.
Data import, as the first step of data analysis, seems simple but very important, especially when you have a large amount of data, you will find these methods particularly useful.
