Importing and loading data in CMA
This article helps in importing and loading data from other programs such as Ms-Excel. Furthermore, it will take up the case example of Cholecystitis and gallbladder carcinoma illustrated in manual data entry procedure. This feature allows to easily transfer data from other programs.Consequently this also facilitates the transfer of data maintained beforehand, not especially for meta-analysis.
For this article, the data loading or importing procedures will utilize the odds ratio data as described in the previous article. The odds ratio data involves the comparison of ‘Case’ and ‘Control’ subject groups. It will indicate the odds of an individual developing carcinoma of the gallbladder as associated with the presence of gallstones.
Importing from Excel to CMA spreadsheet
From here the instructions will be on loading data from other softwares such as Ms-Excel, SPSS or STATA. The present demonstration includes importing data from Ms-Excel. The MS-Excel datasheet as shown below, contains data for the odds ratio taken from the relevant studies.
The odds ratio refers to the statistical range used for assessing the risk of a specific outcome such as a disease as influenced by the presence of particular factor or exposure (Westergren et al. 2001). Now, importing data from other programs into comprehensive meta analysis (CMA) software spreadsheet could be initiated from any of the two options as shown below.
Either select “Import data from another programme” from the welcome dialog box or:
- go to ‘File menu’.
- Click on ‘Import’.
As signified in the image, selecting any of the two options leads to the ‘Import data’ dialog box, which instructs on how to import the data, which is by performing simple copy and pasting commands. It results in the CMA spreadsheet display as shown below.
However, note that this will lead to the pasting of data values only, not the column names as in the excel sheet. Follow the below mentioned steps to facilitate the process of identification.
The “Identify” tab helps in identification of the relevant fields of data according to the pre-designed format offered by CMA.
Identifying study names
The first column “A” contains the study names. Keep column A active and perform the selection:
- Click on ‘Identify’.
- Bring cursor on ‘Column for’.
- Select ‘Study names’.
This selection leads to the confirmation dialog box as shown below.
Identifying columns for effect size data
Following the identification of study names, proceed towards the identification of the effect size data.
- Click on ‘Identify’.
- Bring cursor on ‘Column for‘.
- Select ‘Effect size data’.
This leads to the dialog box for ‘Identify columns for effect size data’, as shown below.
After selecting ‘Next’ select the relevant study design from the dialog box, as shown below. The given dataset concerns the comparison between two groups. Therefore, select “Comparison of two groups, time points or outcomes”.
As the odds ratios from different studies represent the computed effect sizes, perform the following:
- Select ‘Dichotomous’.
- Bring cursor on ‘Computed effect sizes’.
- Select ‘Odds ratios and confidence limits’.
Furthermore, select any of the available formats suiting their study design. The selection of desired data entry format is followed by identification of column containing the appropriate data, as shown below.
- Column B identifies as odds ratio,
- column C as the lower limit,
- column D as upper limit,
- column E identifies as confidence interval.
In this module, confidence interval is reported at 95%, hence the information is entered manually. The software names the identified columns as per their designation and changes the column names from alphabets to the format entry name.
The figure below shows the final spreadsheet, complete with all the data entry columns. The ‘Odds ratio’ and ‘Logs odds ratio’ are the effect size measures, whereas ‘Std. Err.’ and ‘Variance’ constitutes measures of accuracy. The software automatically computes these fields and displays them in relevant columns.