Hello list members,
To come to some conclusion about the number of actual i-mode users, I tried
and combined statistics from a number of sources. First please read the
explanation below, and, above all, remember the German proverb:
"Only trust statistics you have manipulated yourself"
1) Discussing the actual usage of i-mode was difficult due to the lack of
statistical data on this topic. But now, Hakuhodo Institute of Life and
Living has released data on the actual usage of internet-enabled phones for
internet access.
http://www.nikkeibp.asiabiztech.com/wcs/leaf?CID=onair/asabt/fw/126491
2) We already had data from NTT DoCoMo on the number of i-mode capable
handset owners and monthly subscriber numbers for the i-mode service. This
number is now 21+ mln, but we'll use the November number for consistency
with the above Hakuhodo data: 15,410,000.
http://www.nttdocomo.co.jp/i/contract/index.html
3) D2 Communications has some statistics on i-mode owners discriminated by
age group. I believe this data also only refers to the percentage of i-mode
*owners*. Please be carefull to differentiate between *owners* and *users*
in this context. The Hakuhodo research has the only data identifying this
difference.
http://www.d2c.co.jp/user/body02_01.html
http://www.d2c.co.jp/user/body02_02.html
Now, I combined these statistics in the following spreadsheet:
http://www.pikula.com/i-mode_users.xls
... my conclusion is that the number of i-mode *users* adds up to 3,870,689.
This is only for the age group 19~74 though.
Known problems:
1) Age groups had to be combined for Hakuhodo/D2 data. Therefore "~19" is
left out. "19~29" is a combination of the D2 "20~24" and "25~29" group.
"30~49" is a combination of D2: "30~34", "35~39" and 2/3*"40+".
Consequently, "50~74" is 1/3*"40+". Seemed like a good way to split up the
group "40+" by assuming that only 1/3 of this group (owning and i-mode
cellular phone) is over 50. This does not change the concluding number too
much, since i-mode usage numbers over 40 is only a small contribution to the
total number.
2) Data from the Hakuhodo research accounts for *all* internet-enabled
phones in Japan. So it includes i-mode, J-Sky and EZweb. I had to assume
actual usage is the same for all services. Most probably, it is, except for
maybe Tu-ka which has, to my understanding, a lot of young users.
3) There is no indication in the original texts on the size of the sample
group for the Hakuhodo and D2 data. For all we know, Hakuhodo could have
done it on Shibuya square in just 10 minutes. Still, this would give you
1.000.000 valid responses from what I have seen there ;-) Also the D2
statistics are not dated, but this should not be a major problem.
Final remark: Please understand that I know about the coastline-of-Norway
problem, and I have played Soccer with Mandelbrot just last week. This data
is for my own research, and I just thought I would share it with the list.
If you want to see some *really* sick data (for the die-hards out there) I
also found a graphical relationship between new DoCoMo PDC buyers and i-mode
subscribers... Trust me... you don't wanna know ;-)
Regards,
Victor
[ Did you check the archives? http://www.appelsiini.net/keitai-l/ ]
Received on Tue Mar 27 19:08:29 2001