Text Correction in Google Docs

The five-minute video that I've embedded below shows how to use text correction tools in Google Docs. As I mentioned in a previous post, auto-correction functions can help you tidy up texts that you've typed quickly, as you do for typing homework assignments (Using Correction Tools after Typing Trials, March 7, 2013).

Instead of a typed text, for this demonstration, I've started correcting a short passage generated from a graphic by Google's optical character reading (OCR) technology. Yet, as the screenshot below the video indicates, the correction tools don't work perfectly.

Even after five minutes treatment with Google correction tools, a number of errors remained, some still marked by Google (dotted red underscores), but without suitable suggestions for auto-correction. There were others not spotted by Google at all. The latter I highlighted by hand either during the demonstration (yellow backgrounds) or afterwards (blue backgrounds). Those still needed hands-on checking and correction.

To make a long story short, after complete correction, and prettied up a bit for a block quotation, it looks like this:

... A fault line runs through the disciplines concerning culture. On one side are disciplines like history or cultural anthropology, rooted in a historicist logic of seeking local regularities within a bounded milieu. On the other are disciplines like economics, driven by a functionalist logic of seeking transhistorical generalizations. Organizational behavior involves both of these logics.... Yet, the emic and etic perspectives each provide only half of the story. ... / ... [A] richer account of culture can result when an integrative explanatory framework arises. 

(Morris, Leung, Ames, & Lickel, 1999, p. 790)

In retrospect, there seem to have been a number of OCR-generated errors in the passage, for example the two "cx" strings remaining in the second to last line of the first paragraph (one stand alone, and the other in the middle of an underscored word). I had corrected another instance of "cx" to "a" while making the video. If I had used the "Select all matching text" option the first time, I might have been able to correct all three at once.

The string "ol" appears to be another such OCR error, a misreading the word "of", as do the "lo" string, a misreading the word "to", and OCR-generated periods instead of commas (¶1, lines 2 and 7; and ¶2, line 1). For that particular typeface and layout (serif, with full-justification, in the original), Google seemed to have had trouble with commas, and with the letters a, f, and t.

Once you begin to recognize recurring errors, in your typing as well as in optical character read texts, it is possible to use the Find and replace function (Edit menu) to correct numerous errors at once.

For instance, the search shown above would:

• Seek " ol " – with single spaces before and after the letters, to find only stand-alone instances of "ol" – not words like alcohol, oligarchy, or polyphenol; and 
• Replace " ol " with " of " – similarly spaced.

If there are any hits on those search term[s], the buttons across the bottom of the dialog box will become active, and have dark lettered labels.

That fine-tuned search might be a safe bet for the Replace all function (circled in orange, but still grayed out, above). However, if you're not absolutely certain that your search and replacement terms are exact, it will be better to review and replace search strings one at a time using the Next, or Previous, and Replace buttons.

Reference

Morris, Michael W.; Leung, Kwok; Ames, Daniel; & Lickel, Brian. (1999). Views from inside and outside: Integrating emic and etic insights about culture and justice judgement. The Academy of Management Review, 24(4), 781-796. Retrieved November 15, 2012, from http://www.jstor.org/stable/259354

[609 words]