Help in determining age

H

Howard Brazee

I have a spread sheet with the following:

# Name DOB Inauguration date Inauguration age
1 George Washington 2/22/1732 4/30/1789


I input in E2 =D2-C2, expecting to have his age in days - but I get
#Value error.

I actually want age in a format showing him to be 57 years, 2 months &
8 days (maybe 57y2m8d - or some other compact value).

What is my error? Anybody have a good formula for getting me year,
month, and day?

--
"In no part of the constitution is more wisdom to be found,
than in the clause which confides the question of war or peace
to the legislature, and not to the executive department."

- James Madison
 
H

Howard Brazee

I have a spread sheet with the following:

# Name DOB Inauguration date Inauguration age
1 George Washington 2/22/1732 4/30/1789


I input in E2 =D2-C2, expecting to have his age in days - but I get
#Value error.

I see one problem - all of the dates older than 1900 in this spread
sheet are left justified, all of the newer ones are right justified.
It appears that I need to do something special to handle older dates.

Only when both dates are greater than 1900 do I get a subtraction of
dates.

I just don't know the solution.

--
"In no part of the constitution is more wisdom to be found,
than in the clause which confides the question of war or peace
to the legislature, and not to the executive department."

- James Madison
 
R

Rick Rothstein

Given that the Gregorian Calendar (along with its concept of Leap Years)
wasn't adopted in the United States (actually, Great Britain and its
colonies at the time) until 1752, do you know if the functions you pointed
to handle that or not? I ask because more than a third of the time frame the
OP is asking about took place when Leap Years didn't exist.
 
H

Howard Brazee

But John Walkenbach has an addin that may help you:
http://j-walk.com/ss/excel/files/xdate.htm

That looks like my best bet. I have been playing with some other
macros found on the Web.

--
"In no part of the constitution is more wisdom to be found,
than in the clause which confides the question of war or peace
to the legislature, and not to the executive department."

- James Madison
 
P

Peter T

(along with its concept of Leap Years)

The concept of Leap Years was figured out by the Romans and incorporated
into the Julian Calendar. What they got slightly wrong was it gets out of
sync by almost 3 days every 400 years. This was sorted out by Pope Gregory,
with the new calendar incorporated throughout the world starting in the late
c.16th with the last few countries not until well into the c.20th.

Regards,
Peter T
 
R

Rick Rothstein

You are correct (I should learn not to rely on my memory of things I haven't
thought about for years<g>). What I was supposedly thinking about was the
10-day adjustment to the calendar and the new Leap Year calculations that
Pope Gregory came up with... it is that which was adopted by the United
States (colonies) in 1752. My question about whether the referenced link
properly handled this calendar adjustment still stands.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads

Programming question 5
Sending a macro-enabled spreadsheet 2
Months between two dates 5
Lost horizontal scroll bar 0
Ignore error 3
VB macros in new Excel 24
Names 5
Different kind of cell reference. 11

Top