September 21, 2021, 08:07:43 AM
Forum Rules: Read This Before Posting

### Topic: the total percentage error  (Read 12575 times)

0 Members and 1 Guest are viewing this topic.

#### yerbutnobutyer

• Guest
##### the total percentage error
« on: March 28, 2005, 09:53:09 AM »
hello!
i was wondering if anyone knew how to calculate the TOTAL % error. ive just finished a titration and im writing my analysis, ive calculated the individual % errors for each piece of equipment that was used eg:pipette. However i dont know how to calculate the overall % error . Id be really grateful for any help.
thanks sarah x

#### Mitch

• General Chemist
• Sr. Member
• Posts: 5297
• Mole Snacks: +376/-3
• Gender:
• "I bring you peace." -Mr. Burns
##### Re:the total percentage error
« Reply #1 on: March 28, 2005, 10:49:18 AM »
look up something called "error propagation".
Most Common Suggestions I Make on the Forums.
1. Start by writing a balanced chemical equation.
2. Don't confuse thermodynamic stability with chemical reactivity.
3. Forum Supports LaTex

#### Donaldson Tan

• Editor, New Asia Republic
• Retired Staff
• Sr. Member
• Posts: 3178
• Mole Snacks: +261/-13
• Gender:
##### Re:the total percentage error
« Reply #2 on: April 02, 2005, 05:53:16 PM »
given A = B x C, then the error of A is:
dA/A = dB/B + dC/C
"Say you're in a [chemical] plant and there's a snake on the floor. What are you going to do? Call a consultant? Get a meeting together to talk about which color is the snake? Employees should do one thing: walk over there and you step on the friggin� snake." - Jean-Pierre Garnier, CEO of Glaxosmithkline, June 2006

#### mithrilhack

• Guest
##### Re:the total percentage error
« Reply #3 on: April 02, 2005, 11:39:38 PM »
you expect him to do derivatives?

#### Mitch

• General Chemist
• Sr. Member
• Posts: 5297
• Mole Snacks: +376/-3
• Gender:
• "I bring you peace." -Mr. Burns
##### Re:the total percentage error
« Reply #4 on: April 02, 2005, 11:46:37 PM »
You can't do error propagation without derivatives, more generally partial derivatives.
Most Common Suggestions I Make on the Forums.
1. Start by writing a balanced chemical equation.
2. Don't confuse thermodynamic stability with chemical reactivity.
3. Forum Supports LaTex

#### hmx9123

• Retired Staff
• Full Member
• Posts: 897
• Mole Snacks: +59/-18
##### Re:the total percentage error
« Reply #5 on: April 03, 2005, 05:39:00 AM »
If you know what the individual errors are, you can add them up without doing derivatives in most cases.  Unfortunately, it involves some math that I can't recall at the moment.  There's different procedures for additive, divisive, etc. errors, so it depends on how you're manipulating the data.  Look in a statistics book or just search the web for 'additive errors' or 'adding errors'.  It's in my analytical chem book, but that thing is 2100 miles away from me right now (a worthy place for it).

#### Mitch

• General Chemist
• Sr. Member
• Posts: 5297
• Mole Snacks: +376/-3
• Gender:
• "I bring you peace." -Mr. Burns
##### Re:the total percentage error
« Reply #6 on: April 03, 2005, 03:16:34 PM »
You typically just can't add errors though. Additive error addition involves the square root( of the sum of the individual errors individually squared.) But, it all depends on the setup of course.
Most Common Suggestions I Make on the Forums.
1. Start by writing a balanced chemical equation.
2. Don't confuse thermodynamic stability with chemical reactivity.
3. Forum Supports LaTex

#### pkothari13

• Guest
##### Re:the total percentage error
« Reply #7 on: April 03, 2005, 10:31:16 PM »
why not take the average of all your experimental values needed to titrate your solution and then find the percent error from the literature value.

or try standard deviation

hope that kinda helped

#### sammyr

• Guest
##### Re:the total percentage error
« Reply #8 on: April 14, 2005, 04:38:38 PM »
percentage error =       (Error / Reading) x 100

#### arcanineghost

• Guest
##### Re:the total percentage error
« Reply #9 on: April 17, 2005, 09:52:09 AM »
at our high school you calculate the percent error by taking the absolute value of the actual/theoretical value - experimental data all divided by the actual/theoretical value times 100%

Looks like this:

|actual-experimental|
-------------------------   x 100% = experimental percent error
actual

hope that helps