rowSums

rowSums is a function in the R programming language that computes the sum of each row in a matrix or array-like object. It returns a numeric vector with one element for every row, making it useful for creating row totals or per-row aggregates in data analysis. The function is part of base R and is commonly used with numeric data.

The typical usage is rowSums(x, na.rm = FALSE, dims = 1). Here, x should be an object with at least two dimensions, such as a matrix or a two-dimensional array. The na.rm argument controls how missing values are handled: if na.rm is FALSE (the default), any row containing an NA yields NA in the result; if na.rm is TRUE, NA values are ignored when computing the row sums. The dims parameter is used for higher-dimensional arrays and is less often needed for simple matrices.

Notes on data types and preparation: rowSums operates on numeric data. When used with data frames, it is common to select numeric columns or convert the frame to a numeric matrix first to avoid unintended coercion of non-numeric columns.

Examples: For a 2x3 matrix m created by filling values 1 through 6, rowSums(m) returns c(6, 15). If m contains missing values, such as m <- matrix(c(1, NA, 3, 4, 5, NA), nrow = 2, byrow = TRUE), rowSums(m, na.rm = TRUE) yields c(4, 9). With a data frame like df <- data.frame(a = c(1, 2), b = c(3, 4)), rowSums(df) returns c(4, 6).

See also: rowMeans, apply, colSums. Similar functionality exists in other languages, such as NumPy’s sum with axis=1 in Python and pandas’ sum(axis=1).