Could use help designing small database :)

[vanekerenam]

Hello everyone,

for a little exercise I've been asked to make a small 'fake' database (only design tables + create relationships) that 'could have been used' for a certain paper.
It would be amazing if some of you could have a look and then get back to me whether my database is well-designed, and where I could have made errors, and perhaps where there are improvements possible.
Here is the link to my database:
fast-files.com/getfile.aspx?file=129366

Below I'll give the info I can work with to make the database:


it's about the monitoring of populations of two species of albatross at different locations. They did this by counting nests containing one or more eggs, banding juveniles and adults, and by counting recaptured banded birds.
"Monitoring of the wandering albatross population was
carried out on Possession Island, one of the
Crozet islands in the southwestern Indian Ocean (Fig. 1),
between 1960 and 1995. The entire population of Possession
Island was censused seven times between 1960
and 1980, and every year since 1981. Three colonies are
found on the island, at Baie du Marin, Est and Pointe
Basse (Weimerskirch & Jouventin, 1987, for location).
The number of nests with an egg was recorded when all
laying had been completed, i.e. around 15 January every
year. A total of 3312 adult birds and 4750 chicks prior
to fledging were banded with monel bands between 1960
and 1994. Since 1966, recapture of banded birds has
been undertaken each year between December and
April. Bands of breeding birds were checked after egglaying
until April, bands of non-breeders between
December and April. At Kerguelen, wandering albatrosses
were censused in mid-January on the Courbet
Peninsula in 1971 and between 1985 and 1993 (see
Weimerskirch et al., 1989a). The small Amsterdam
albatross population was censused on Amsterdam
Island every year between 1983 and 1994, and breeding
success was determined from monthly checks of nests
from egg-laying in March until fledging the following
January. Since 1983, all chicks have been banded with
monel and plastic bands prior to fledging; recording of
banded birds is carried out each year between March
and May.
" (Weimerskirch, Brothers and Jouventin, 1997)
Thank you very much in advance!

 

[jdraw]

cross posted at http://www.utteraccess.com/forum/index.php?showtopic=2041235&hl=

Your link is not accessible to me.

Is this some sort of homework/assignment?

This may be helpful and I recommend you post directly to the forum.

 

[vanekerenam]

Well it should be accessible if you put www(dot) in front of it.
could also try 321webs.com/download/129366.htm

The answer to your latter question is affirmative!

 

[vanekerenam]

try to put www (dot) in front, that might help , if not
321webs.com/download/129366.htm

answer to the last question is yes

 

[Frothingslosh]

I definitely smell homework.

Ironically, I was just thinking today about how it had been a while since I'd seen a 'please fix my homework' post here.

 

[vanekerenam]

Well I'm not looking for someone to do my homework for me, I'm just asking for someone to check it and perhaps give me some advice. Don't know why it wouldn't be allowed to ask for help with your education and it would be allowed to ask for help with your job.

 

[Frothingslosh]

Most of us are unwilling to download files from third-party sources. Follow the instructions on jdraw's post to put up a zipped copy of your database and someone will probably take a look at it. (I won't be able to because I'll be at work, and won't be able to download things there.)

People check to see if it's homework because we tend to treat homework differently - instead of just giving answers, we tend to point people in the direction to figure it out themselves unless those answers are unusually esoteric.

Also, THIS link is something you should probably read for future posting.

 

[vanekerenam]

Thank you for the advice, I have now attached the zip file!

 

[vanekerenam]

Thank you so much for your help,

It is indeed a bit confusing,

Overall there are indeed 5 sampling stations (or if you will, colonies):
Amsterdam island (indeed, only amsterdamensis lives there)
Possession Island, which is one of the Crozet island, has three sampling stations (Est, Point Basse, Baie du Marin),
the last station is on the island called Kerguelen, more specifically on the Courbet peninsula.

Overall, I think the gist of it is just that there are 5 sampling stations, whether these are located on larger islands (one of the Crozet islands,...) is just extra information that not need to be included. For example, I have called one of the stations Kerguelen, because the Courbet peninsula is the only station on that island anyways, on Possession Island more distinctions needed to be made because it has 3 stations.

Amsterdam indeed only seems to have monthly counts.

there is not much detailed info , so I'm afraid I can't acquire starting values for the other species that are not given in the text provided.

 

[vanekerenam]

The aim is to simulate a dB that could be the one these researchers used during this project, which nicely relates all data to each other that they collected (even though some information might not be in the text provided). Hence the databse i had so far with 5 tables)

 

[vanekerenam]

The dB i have to develop only incorporates things that have happened until the point this paper was published, the dB might need to be tweaked if this'd happen, but since it is not included in the text I was provided with it's "not my concern"

 

[Frothingslosh]

Well, specifics will depend on how your teacher/professor handles going above-and-beyond, but in the real world, it's important to future-proof a db as much as possible, as long as it doesn't put you over budget or cause delays (and sometimes even then).

The rule of thumb is that if there is a possible situation that can screw up your database, then it WILL happen, and it needs to be accounted for. If you don't do it during design, then you will absolutely be doing it after release, and that is almost always more difficult.

That said, we're aware classroom environments tend to be a little more limited in scope, and some professors/teachers really dislike people going beyond the designated limits of the project.

Some bosses too, for that matter.

 

[Frothingslosh]

should that be censi

'Censuses' in English, based on both my own experience and everything I could look up. (I checked because I got burned not too long ago on 'temperament' - the 'a' isn't pronounced in the Great Lakes accent, so I didn't even realize it was there!)

It DOES seem kind of random when we use the Latin pluralization (which AWF doesn't consider to be a word, it seems), and when we go with the English rules. In this case I think it's because it's a created meaning attached to what was originally the past participle of 'censere', which means 'to assess'. Apparently English rules apply because it wasn't a noun in Latin.

/derail

 

[vanekerenam]

Gizmo, thank you so much for the help you've been already,

I think that, because this is an assignment to make us familiar with databases, I should not worry about how to treat the collected data to make them fit into a database.
All that has been asked of me is to make a dummy dB that the researchers of this project could have used (even if it's oversimplified), hence I linked the dB I made to see whether people here thought it made sense a bit, or whether there are mistakes in the relationships, things incorrect in the tables etc...

Thanks!

 

[vanekerenam]

so, could anyone check the database I made and see whether it makes sense in regards to the text provided?
It'd be much appreciated!

 

[Frothingslosh]

Ah, I thought Gizmo already had. I'll try to remember to do so after work, but I won't be home until after 5 (it's 7:26 now).

 

[The_Doc_Man]

vanekerenam,

Since you are still learning, here is my advice to you, some simple guidelines.

We have some workman's aphorisms here for building things. One of them for carpenters is "measure twice, cut once." The database equivalent is my Old Programmer's Rule #1 - "You can't do it in Access until you can do it on paper." Building a database "as you go" is (sadly) a commonplace activity, sometimes forced by circumstances. However, taking the time ahead of time to analyze the problem for elements that will need to be stored is incredibly important.

During this analysis, you run into the second Old Programmer's Rule: "Access won't tell you anything you didn't tell it first." That is, if you want X to be in a report, it has to be in the database for the report to be able to get it. Therefore, sometimes you have to work backwards from goal to data inputs in order to know what you need to capture in order to get the results you want.

Finally, one VERY simple principle will ALWAYS be correct (unless your teacher is trying to teach you some type of oddball workaround.) The closest English aphorism is "Don't let the tail wag the dog." That is to say, when you have a difference between your design and your data, your design is usually wrong because the data is (are) always right.

 

[vanekerenam]

Thanks a lot for your advice Doc Man, I'm sure it will come in as a great help when I need to build a dataset myself

However, I was wondering if anyone could just take a quick look at the database I made, and see whether it makes sense in regard to the text I provided in my initial post. Whether the relationships make sense etc. You'll see the database is very basic, I didn't even put in any actual data, it's just about creating a more or less usable format.

I'll attach the database again quickly !