As the Wikipedia article on that quote explains, Disraeli most likely wasn't the source of it and the original source is unknown.
Jeremy Miles and Philip Banyard explain in Understanding and Using Statistics in Psychology: A Practical Introduction (p.3, n. 1) that Mark Twain found that quote in an essay by Leonard Henry Courtney in The National Review in 1895. Courtney wrote:
After all, facts are facts, and although we may quote one to another with a chuckle, the words of the Wise Statesman, "Lies - damned lies - and statistics," still there are some easy figures the simplest must understand, and the astutest cannot wriggle out of.
Miles and Banyard write that
Twain thought that Courtney was quoting Disraeli when he wrote 'the Wise Statesman', but Courtney was referring to a hypothetical wise statesman, not a specific one. Rather spoiling the whole quote, it has been suggested that the dashes are parenthetical, and Courtney was trying to say something like 'Lies (damned lies!) and statistics'.
This means that when you want to quote that phrasing of this saying you must cite Twain, not Disraeli, as Twain is the source of it. Depending on what the purpose of the quote is within your text, you could either write:
"There are three kinds of lies: lies, damned lies, and statistics." (Twain, 1907, p. 471)
or something like:
Mark Twain (1907, p. 471), who attributed the quote to Benjamin Disraeli, wrote: "There are three kinds of lies: lies, damned lies, and statistics."
There is no need at all to quote from the Oxford Dictionary of Quotations, as the original publication is available online for free.
Following APA, your reference would look like this:
Twain, M. (1907). Chapters from my autobiography – XX. The North American Review, 185(618), 465–474. https://www.jstor.org/stable/25105919
You must always attempt to find the primary source. Today you can obtain most books at every university library worldwide through interlibrary loan. Having to pay for that service is no excuse.
If Twain's work was actually unavailable to you, you would list only the secondary source (the Oxford Dictionary of Quotations) in your reference list:
Knowles, E. (Ed.). (1999). The Oxford dictionary of quotations. Oxford University Press.
In text you would identify the primary source (Twain) and write "as cited in" the secondary source that you used, e.g.:
... (Twain, 1907, as cited in Knowles, 1999, p. [page number])
Tertiary sources – that is, information compiled from other sources such as textbooks, encyclopedias, or dictionaries – are not allowed in academic papers, unless their study is the purpose of your writing.
In their latest Q&A, the Chicago Manual of Stile answered a related question:
Q. If an author insists on using a widely attributed quote whose source cannot be confirmed, how should I cite it? Should I cite it at all? Or should I simply note in the running text that the quote is “widely attributed to such and such”?
A. Your last idea is the best one. In general, if the source of a quotation can’t be confirmed, this fact should be stated in the text that introduces it instead of being relegated to a note, where readers might miss it. The phrase “widely attributed” should make it clear that the attribution isn’t definitive; however, if the author can provide one or more sources that back up this claim, those could be (and in an academic work should be) provided in a note.
