Making subfields

[Clarendon]

Strangely enough, my question in the Database design keeps disappearing and I
don't know why. So I am posting it again here in the General.

I am working with data that seem to require subfields. I am using Access 07.
I am analyzing features of instances of phrases I search in the Internet:

Phrase Category 1 Cat 2 Cat 3
A 5 / 50% 5 / 50% 0
X2, Y3 Z5
B

The table shows that phrase A has 10 instances, 5 in Cat 1 and 5 in Cat 2.
These instances have features marked with XYZ. Table shows that for A the
five instances of Cat 1 are broken down to 2 of X and 3 of Y, and so forth. I
would like to sort this data like: 'for A list the features that are in the
Categories that score freqencies below 10%' and 'list the Categories with
only one feature (Cat 2 in the table)'.

Can access handle this data? I think multi tables is not helpful because
they only add further info for values, not fields. I need to add further info
for fields (Cat) by setting up subfields of frequency and feature. Also, I
don't think multivalue option will help because it works only as a lookup
(can't add random numbers).

I tried using two primary keys (Phrase and Type) to correspond to each
subfield as follows:

Phrase Type Cat 1 Cat 2 Cat 3
A Freq. 5 5
A Perc. 50 50
A Feature X2, Y3 Z5
B
B
B

This may be a solution, but is quite cumbersom. Also, I am not sure if I
will be able to do the queries I mentioned above. Thank you very much for
your help.

 

[Bob Quintal]

Strangely enough, my question in the Database design keeps
disappearing and I don't know why. So I am posting it again here
in the General.

I am working with data that seem to require subfields. I am using
Access 07. I am analyzing features of instances of phrases I
search in the Internet:

Phrase Category 1 Cat 2 Cat 3
A 5 / 50% 5 / 50% 0
X2, Y3 Z5
B

The table shows that phrase A has 10 instances, 5 in Cat 1 and 5
in Cat 2. These instances have features marked with XYZ. Table
shows that for A the five instances of Cat 1 are broken down to 2
of X and 3 of Y, and so forth. I would like to sort this data
like: 'for A list the features that are in the Categories that
score freqencies below 10%' and 'list the Categories with only one
feature (Cat 2 in the table)'.

Can access handle this data? I think multi tables is not helpful
because they only add further info for values, not fields. I need
to add further info for fields (Cat) by setting up subfields of
frequency and feature. Also, I don't think multivalue option will
help because it works only as a lookup (can't add random numbers).

I tried using two primary keys (Phrase and Type) to correspond to
each subfield as follows:

Phrase Type Cat 1 Cat 2 Cat 3
A Freq. 5 5
A Perc. 50 50
A Feature X2, Y3 Z5
B
B
B

This may be a solution, but is quite cumbersom. Also, I am not
sure if I will be able to do the queries I mentioned above. Thank
you very much for your help.

Proper relational database design would require a 3 field primary key

Phrase
Type
Cat
and the value
Giving
Phrase Type Cat Value
A Freq. 1 5
A Freq. 2 5
A Ftre 1 x2
A Ftre 2 Y3
A Ftre 3 Z5

you don't need percentage, it can be calculated from the frequencies.