NCA Software and Calculator

Version 4.0.1 is available

From version 4.0.0 the NCA software in R includes two new functions. The first is nca_random that generates a random dataset that meets a necessary condition with a given straight ceiling line. The function can be used for simulations with NCA. The second new function is nca_power. This function estimates the power of NCA's statistical test, which is the pre-study probability that the test will reject the null hypothesis when necessity exists. The function can be used to determine the minimum required sample size. Details of both functions can be found in the online book.

If you have installed an older version of the NCA package you can update the package as follows:

update.packages("NCA")

The NCA package uses other packages. It is possible that these packages are not installed on your computer. In that case the following error message may appear: 'There is no package called …' , where … corresponds to the name of the missing package. You then need to install the specific packages first.

install.packages("…")

If you install the package for the first time: See the Quick Start Guide in the downloads below. A novice user without knowledge of R or NCA can install the software on the user’s computer and perform an NCA analysis within 15 minutes.

The Quick Start Guide and demonstration show how NCA can be done in three steps (both can be found in the downloads below).

The software draws ceiling lines and produces several NCA parameters (including necessity effect size). An example is shown here.

A video explanation of the bottleneck table , and a video demonstration on how to produce the bottleneck table with the NCA software are available.

First Stata version is available

You can find the quick start guide for conducting NCA with Stata in the downloads below.

The effect size d can have values between 0 (no effect) and 1 (maximum effect). An effect size of 0 < d < 0.1 can be considered as a ‘small effect,’ 0.1 ≤ d < 0.3 as a ‘medium effect,’ 0.3 ≤ d < 0.5 as a ‘large effect,’ and d ≥ 0.5 as a ‘very large effect’.(Dul, 2016, p.30).

Prepare a CSV file where rows correspond to cases, and columns correspond to the X and Y variables. The calculator only works with CSV files using a comma ( , ) as delimiter, and a period ( . ) as separator. If needed, the delimiter and separator can be changed in Excel. With files that do not align with this format, the calculator will return an error message. 

 The NCA calculator is here. For help, see below.

An example of a bivariate necessary condition analysis is shown below. For further analyses (multiple NCA, additional NCA parameters, etc.) you can use the NCA software (see above).

Help for calculator

The NCA Calculator calculates the necessary condition effect size and displays the scatter plot with a ceiling line. The calculator can be used in three steps:

1. Select options. First, decide which ceiling line will be displayed in the scatter plot. If your data are discrete (the variables have two or more levels) select the option "discrete". The calculator will display the CE-FDH ceiling line (step function). If your data are continuous (the variables have a large number of levels up to infinity) select the option "continuous". The calculator will display the CR-FDH ceiling line (straight line). When the number of observations is small, you may consider to use the "discrete" option, even for continuous variables. Second, decide if you also want to display the Ordinariy Least Squares (OLS) regression line through the middle of the data, for reference.
2. Enter data. By default the data window displays the data from an example, where Condition X is a country's Individualism and Outcome Y is a country's Innovation Performance. Each row is a case (country). For using your own data prepare a CSV file where rows correspond to cases, and columns correspond to the X and Y variables. The first row should contain headers (variable names). The calculator only works with CSV files using a comma ( , ) as delimiter, and a period ( . ) as separator. If needed, the delimiter and separator can be changed in Excel. With files that do not align with this format, the calculator will return an error message. Upload your CSV file (Browse). When the data is uploaded a window appears where you can select your X variable (the condition) and your Y variable (the outcome) by entering the corresponding column number. The variable name appears in the box below the column number, and can be changed (this name is maintained during a calculator session; if you wish to recover the original variable name you must re-upload the CSV file). The variable names will be displayed at the X and Y axes of the scatter plot (see step 3). You can select the box "Use first columns for row names" to display row names in the data window. Select OK to show the new data in the data window.
3. View results. Press the "View results"  button to display the calculated effect size (d) and the scatter plot with the selected ceiling line and the OLS line if selected. The effect size can be compared with benchmark values. The effect size is one major NCA parameter for evaluating necessary conditions. Other parameters can be calculated with the NCA Software.

You can repeat the analysis with other variables by clicking "Change X and Y selection" under "Enter data"  

The NCA calculator is here.