Method for effective proof reading and error search?

Aharon · Post by **Aharon** » Fri May 11, 2012 5:53 pm

I need advice on proof reading. Specifically:

How can I effectively read through 10-20 page reports, searching for spelling, formatting and similar mistakes?

and, more importantly, how can I effectively check calculations and tables done in excel for errors?

What I'm looking for is some kind of method to do those tasks. Currently, I try to check my results, but it is hard for me not just to glaze over the finished work - I'm familiar with it and it is hard for me to read a familiar text/table/calculation thoroughly.

Does anybody know how one can improve in this respect?

RobbyPants · Post by **RobbyPants** » Fri May 11, 2012 6:11 pm

Well, as a general rule, when it comes to proof reading, I'd say try to finish it early enough so you can at least let it sit overnight before you read it again.

As for checking calculations, if this is in a table, couldn't you just use Excel or something similar to do the calculations? At that point, all you have to do is make sure your formula is right and that you entered your numbers correctly.

Ancient History · Post by **Ancient History** » Fri May 11, 2012 6:15 pm

The simple answer is "give it to someone else." Proofing your own work can be difficult because you're too close to it. If you are going to proof your own reports, then set it down for a couple hours or days and then go back and read it from the beginning.

Re: Excel, the best way to do things is to have a test column/cell where you put in known values and see if the output is what you expect.

Aharon · Post by **Aharon** » Fri May 11, 2012 6:22 pm

Thanks to both of you!

@RobbyPants
=> Let it sit overnight
Good advice, I will try to do that more consistently
=> checking calculations
Making sure my formula is right and I entered the numbers correctly is the problem I wish to solve. Recently, I overlooked that a formula I inserted that simply added two numbers didn't add the correct numbers - one reference was to the wrong column. The result wasn't immediately obvious because it still seemed plausible. I look for a way to avoid that kind of mistake.

@Ancient History
=> give it to someone else
Possible, but that reflects on my performance if they do indeed find mistakes I could have corrected. The goal is to eliminate most of the stuff myself so I don't waste my co-workers time.
=> Test column/cell
Could you give a short example, please?

Whatever · Post by **Whatever** » Fri May 11, 2012 6:30 pm

Don't read stuff from beginning to end if you're proofing. At minimum, start at the end and read to the beginning. Read pages and paragraphs out of order. Make sure you look at everything (ideally not on the same day that you've written it), but breaking it up helps you focus on looking for errors.

RobbyPants · Post by **RobbyPants** » Fri May 11, 2012 6:31 pm

Well, you can always proof it yourself before turning it over for a second proof. Still, AH is right in that you'll likely get more errors caught if someone else looks at it.

Ancient History · Post by **Ancient History** » Fri May 11, 2012 6:38 pm

Aharon wrote: @Ancient History
=> give it to someone else
Possible, but that reflects on my performance if they do indeed find mistakes I could have corrected. The goal is to eliminate most of the stuff myself so I don't waste my co-workers time.

Review time is something you build into the process: somebody reviews your stuff, you review someone else's stuff. You might suggest to the boss to add in a proofing phase, at least experimentally, see how other people in your workplace like it.

=> Test column/cell
Could you give a short example, please?

Okay, let's say you've got a bunch of experimental data in columns A - Z, with each column representing data from a different device and you've set up the spreadsheet to put the data in each column through a couple different formulas to translate the raw data into different units, and then process that data to pop out the Mean, Median, and Mode in different cells (A27, A28, and A29 for column A, for example).

To check yourself, you make a test column (ZZ) filled with expected data - 111222333, that kind of thing - and put it through the same transformation and processing. If the output in ZZ27, ZZ28, and ZZ29 doesn't match your expectations for data when you know the input, then there's an error in your process and you need to go back and check your formulas.

Aharon · Post by **Aharon** » Fri May 11, 2012 7:00 pm

@RobbyPants
It looks like I'll have to do it this way, then.

@AH
Thanks for the excel example!

Some more background: We're a small enterprise (boss, six employees, secretary, two trainees). Except our secretary and the trainees, everybody has an academic degree. We did try to institute that as a rule, but only me and one person working from home office consistently do so. The person working from home office is also, at the moment, very busy because of some deadlines, so I can't ask him to proofread. The others do proofread, but don't ask for proofreading in return, which makes asking a bit awkward. They are either better at proofreading their own stuff than I am or make less mistakes in the first place.

In fact, I fear the underlying problem is that I am not able to concentrate as well as some of my colleagues do, so my work is more error prone. Making as little mistakes as possible in the first place would obviously be the best solution, but I have even less of an idea how to achieve that, given my current abilities and work conditions.

Maj · Post by **Maj** » Fri May 11, 2012 9:28 pm

I do the best proofreading when I read out loud.

shadzar · Post by **shadzar** » Fri May 11, 2012 11:08 pm

for spellchecking, put it in as plain a text writing program you can, that can check spelling.

then have the computer read it to you with a text-to-voice engine. this will let you catch words that may be spelled right, but are the wrong word due to transposition of letters.

know the common homophone mistakes:

to/too
there/their
your/you're
etc

RiotGearEpsilon · Post by **RiotGearEpsilon** » Mon May 14, 2012 8:24 pm

Drastically change the font, color, and layout.

K · Post by K » Mon May 14, 2012 11:26 pm

Printing things out is a great way to catch mistakes. Don't ask me why it works, but it does.

Popping things in Word is also a great way to catch the easy mistakes, but it's crap about words that sound the same but are different.

Finally, letting things sit is great. It usually take 3-5 days for your short-term memory to lose something, so you can look at it with fresh eyes.

fectin · Post by **fectin** » Tue May 15, 2012 8:25 pm

Mentally diagram long sentences.

Columns are easier to proofread, and can look better too.

CatharzGodfoot · Post by **CatharzGodfoot** » Tue May 15, 2012 9:30 pm

Aharon wrote:In fact, I fear the underlying problem is that I am not able to concentrate as well as some of my colleagues do, so my work is more error prone. Making as little mistakes as possible in the first place would obviously be the best solution, but I have even less of an idea how to achieve that, given my current abilities and work conditions.

Get diagnosed with and medicated for attentive-primary ADHD.

JigokuBosatsu · Post by **JigokuBosatsu** » Thu May 17, 2012 6:01 pm

Definitely read it backwards/out of order.