Guidelines for Better Results

Okay, just to get this into the open--I think too much. So bleah. Deal with it. I know a bit about basic statistics, and looking around, some of these experiments just aren't up to par. These guidelines will improve the quality of the output of the team significantly.

First of all--Since a lot of these studies involve dream recall, I need 25 people to FAITHFULLY record how many dreams they remember each night for two weeks.

Second, on the same idea--I need a second, different group of another 25 people to FAITHFULLY record how many lucid dreams they have per month.

If possible, I will be PMing 50 randomly selected members (again, to eliminate bias) to do this. However, since not all of our 2,700-odd members are active or are able to participate, this may be difficult. Anyway. Onward and upward.

Now, as soon as this is done we can get an approximate idea of the mean number of recalled dreams and LD that a user on DV has. This will help significantly later.

After this, we can give far more accurate results on experiments. If any of you who are already working on an experiment want help on making it more realistic and reliable, just PM or email me any time and I'll gladly help.

Also, just a note, as a basic guideline, choose at least 25 people for any experiments involving LD counts or recall counts. It's just better that way.

You are right. But I feel the members are doing a fine job. It is hard to follow a specific format when at the same time you are trying to pave the way into uncharted territory.

I agree with your Lding account. I keep an accurate journal of my Lucid dreams, when I have them, what position I was in, and so forth. So I would be willing to be one of the participants in your study. (The Lucid dreaming aspect of it)

I agree, it is somewhat unpaved territorry. But we need to standardize some things so that results don't vary as much. The 50 people who get chosen will be the universal 'control group'. Then, all results can be compared to them and it can be determined whether or not the technique in question (the hypothesis) is correct, and if it is, how much it affects things.

You're very right about how well the members have done so far. They've shown great ability to make sensible experiments, but this way we can directly compare multiple people's results to give more in-depth analysises of results.

Also, the people have to be randomly chosen (which is going to be a pain in the @$$ for me, since obviously not everyone on the site is active), so I can't accept volunteers. This way there's not any bias. For example, if I just asked for 25 volunteers, they'd most likely be moderately to highly active members who've practiced dream recall and LD techniques and therefore achieve more LD's and remember more dreams that the true mean. Thanks for the offer, though, it's good to know that I have support!

that's okay, I'm a moderately active member that talks about LD'ing every day but never does any pratice whatsoever!!!!!!!!!!! YAY!!!!!!!!!!!!!! (okay off topicish). i have had no lds for the past 4 months and up.

But you are right. Some organization is required.

[quote]Also, the people have to be randomly chosen (which is going to be a pain in the @$$ for me, since obviously not everyone on the site is active), so I can't accept volunteers.

I think it would be resonable to start with the Research Team Members.

Do you mean ask them for help, or use them for the experiment? They could be very helpful in contacting and helping collect and compile the data, so I think I'll PM a few of them tonight to ask for help.

Last night I got about halfway through contacting the first control group, but 15 of the 25 weren't very active, so I'm PMing more random people in case they don't respond.

Quote:

---

Originally posted by Tsen
Do you mean ask them for help, or use them for the experiment? They could be very helpful in contacting and helping collect and compile the data, so I think I'll PM a few of them tonight to ask for help.

Last night I got about halfway through contacting the first control group, but 15 of the 25 weren't very active, so I'm PMing more random people in case they don't respond.

---

I meant use them for the expeiriment.- But ya. By all means, if they are a member there is know reason they could not help you compile data. Nice idea!

You may not get all the people that you wish for with just the Research Members, but to start there was my implication. :D

Okay, I'm making a call for help to the research team.

I've come down with something, and I feel roughly like somebody's been shoving steel wool laced with Ex-Lax down my throat. So, for the moment I'm not particularly up to doing much of anything, especially thinking. WAY too hard right now.

So, I'm going to have one of you find the data for me (provided somebody can and will). I need somebody with a graphing calculator (preferrably a TI-83, 84 or 86) and an hour or two of spare time. If you do, please post here and follow these instructions. If somebody else has already replied and is working on it, please don't go on yourself. We only want one person working on it, so that people don't think we're spamming them.

Now, if you've accepted the previous terms and nobody else has, post and notify everyone else that you will start working on it. Then, using your calculator use the RandInt function;
-On the TI-86, go to catalog, then scroll until you find it (It's alphabetical, so look under R)
-On the TI-83, it should be under Math, then Stats, but I'm not 100% on this since I don't use an 83
-On the TI-84, exactly the same as an 83

After you've selected RandInt, your screen will say RantInt(
Enter this: 1,2715,20)
Now, your screen should look like this:
RandInt(1,2715,20)
Press ENTER and 20 random numbers should appear on your screen.

Now, on your computer, go to the memberlist (sort it by joined date if it isn't already) and look up each number (this is the long part) and PM the owner, then PM me their screen name so I know who to expect to hear from. If they've never visited the forum, or haven't in a few months, don't bother PMing them, just go to the next random person.

Afterwards, for every person that you didn't PM (the inactive sort), you'll need to choose more random people so that we have enough for the experiment. To do this, put into your calculator RandInt(1,2715,X) where X is the number of people who weren't active.

When you PM them, write something like:
Hi, I'm so-and-so and I'm helping with an experiment to determine the mean amount of dream recall for DreamViews members.
(If you want, post a link to this thread so they understand entirely)
The experiment will require you to write down how many dreams you remember every night for 2 weeks, most likely beggining March 7 and going to March 21. If you will be unavailable during this period, or don't want to participate, PM Tsen (or just have them reply back to you and forward it to me) and let him know so that we can get somebody else. If you will be available and want to participate, PM Tsen (or as mentioned earlier, PM you and forward it to me) and let us know.
Thanks,
-Your name here

Since it would be a pain to write a new letter to every person, just write one, then copy it and paste it when you send the next one.

If you're willing to help, but don't have too much time, just post that you will do the first 5 or 10 or whatever and somebody else can do the rest. If you do this, make sure only one of you selects the random numbers and PMs the other one of you the first 5/10/whatever random numbers and does the others him/herself. Basically, make sure you don't PM the same person twice (or heaven forbid three times), lest they seek revenge against you (or, even worse, me).

Also, I've already PM'd a few people who have already replied, so don't PM any of the following:

Lomebririon
MeusOpusMagnus
Stalker
DooleyClovis

The rest of my first group hasn't replied yet, so I won't bother posting their names.
Have fun! Well...try. And if nobody replies, I'll just do it when I'm feeling better, but that might delay things a bit. Sorry to dump this much on you guys!

Okay people, I can't say I expected results. I mean, it's a lot of (dull) work, so I'll just take care of it later. I'll be out of town until Sunday, but I am feeling much better now, so I'll be back on it Sunday, maybe Monday night. Catch ya' all later!
-Tsen