You can sort your results by

1. Rank, which lists the hits in order of "most likely candidates" first. The ranks or Relevancy Scores increase the more often a term appears on a page or if it appears in the page title. Items of the same rank are listed alphabetically by page title.
2. Rank, with similar items grouped, the default, which groups pages with other similar pages and then lists the hits in order of the most likely candidate for the group. For example, if your search results in 12 pages from Bentham's Defence of Usury, they will be collected together under the highest ranking page from that book.
3. Rank, concise (omit similar items), which is the same as Rank, with similar items grouped, except that only the first item in each group is reported. Thus, only the highest ranking page from Bentham's Defence of Usury would be listed, along with a notice saying that there are more items like that. Recommended if you are searching for a common term in a book. Not recommended if you are looking for a common term in an article or in the Concise Encyclopedia of Economics.
4. Page Title, which lists the hits alphabetically by the page titles.
5. Page URL, which lists the hits alphabetically by the page URL.

Examples: comparative or advantage finds pages with either the word "comparative" or the word "advantage" (or both words, not necessarily successively or close together). comparative not relations finds pages with the word "comparative" but not the word "relations".
The term not may not work properly if you start your search terms with this word or if you continue on to search a particular book. For best initial results, type bank not central rather than not central bank to find pages with the word "bank" but not the word "central".

Exact phrases. Use quotation marks to search for exact phrases (terms that occur contiguously).

Example: "comparative advantage" finds pages with the exact phrase "comparative advantage". It does not find pages with the phrases "advantage that is comparative" or "comparatively advantageous".

Wildcards: Site Searches ordinarily find only complete words. This can sometimes result in finding 0 hits when there are in fact instances of words you are looking for on a page. To find partial words or words with alternate endings, use an asterisk * at the ends of words. (Note: Asterisk wildcards cannot be used within words in the Site Search tool at the present time. They are only effective at finding alternate word endings.) See also punctuation and ligatures.

Examples: If you are looking for references to Montesquieu, the Site Search tool will report 0 hits for pages where the only use of the term is "Montesquieu's". However, Montesquieu* finds Montesquieu and Montesquieu's; margin* finds margin, marginal, marginalism, marginalist, etc.

Punctuation. Some punctuation is ignored, but other punctuation (including sometimes periods, apostrophes, quotation marks, and hyphens) may cause occasional no-matches to be found even if your terms do have matches. If you get no matches or not as many matches as you expect, remove any punctuation and try again. See also wildcards.

Example: Searching for self interest and self-interest yield different results: the latter is, as expected, a subset of the former. Searching for "self interest", and "self-interest", yield the same results, but fewer results than self-interest without the quotation marks! I don't know all the explanations for these apparently illogical results; so my suggestion is that if you are using punctuation and not finding what you are sure must be there, try eliminating the punctuation.

Case sensitivity. Site Searches are case-insensitive at the present time. (If you are searching a specific book, case may be specified when doing a Book Search.)

Ligatures, special characters, and accented characters may be used directly in the form. The equivalents are automatically matched in most cases. If you want a list of the equivalences, or if for some reason you want to type them in, you can find these characters to copy and paste in this ASCII/HTML codes chart in the leftmost, "Character", column. (The columns entitled "ASCII/ISO Latin-1 character codes" and "HTML entities" will probably not retrieve the expected material in the Site Search form.) Position your cursor over the item in the "Character" column, and then copy and paste it right into your form. You can probably also paste these characters successfully directly from your word processor or any browser application: if it shows up at all in your browser in the input box you paste or type into, it is probably the character you intend. For more on this topic, see Search Books.

Examples: Searching for resume and resumé yield the same results, as do searching for Egypt and Ægypt. However, searching for manoeuvre and manœuvre yield different results. This is a trial-and-error matter.

Extraneous URLs. Site Search Results sometimes include hits to pages that do not appear to contain your search terms. One reason this may occur is that some underlying HTML code is also searched. Consequently, embedded "tags" we've inserted into the text are searched, as are the page headings and meta-text. Another main reason may be that if you are searching for multiple terms without using quotation marks (see exact phrases), your terms may appear far apart on any given page instead of consecutively. Yet another, somewhat rare, reason may be that if you are doing a Book Search following your Site Search, the term you intended to search for may not be passed between the two search engines exactly as you intended it. Try correcting it in the Book Search form; or check out the Book Search syntax for more details and ideas. Also: please check your spelling!

A file's rank (for a single word) is basically the log of word frequency (how many times that word is found in a document) scaled to 1000.

Tech tip: The word frequency is biased by the "structure" which is a flag that says where in the html the word is found (e.g., in <title> or <h1>). For example, a word in the title might be ranked as equal to, say, five occurrences in the body. [—Bill Moseley, Swish-e mailing list, titled "[SWISH-E] MetaNamesRank" Dec. 1, 2002]

The Site Search engine is . Interactivity with the Book Search and Card Catalog Search engines with Swish-E added by Lauren Landsburg.

Copyright ©: 1999-2004, Liberty Fund, Inc.
The cuneiform inscription in the logo is the earliest-known written appearance of the word "freedom" (amagi), or "liberty." It is taken from a clay document written about 2300 B.C. in the Sumerian city-state of Lagash.
The URL for this site is: http://www.econlib.org. Please direct questions or comments about the website to webmaster@econlib.org.