Personal experience is not evidence? The hell you say!

The BBQ Pit
41

…and you’ve learned one thing: that a lot of people are telling similar anecdotes. (‘Rigorously’ assembling unsubstantiated stories doesn’t make the stories themselves rigorous; are trying to be tricky?) Presuming that these remain as ‘mere’ anecdotes (ie: lacking verification as to their veracity) the only further conclusion that you may draw from this accumulation of anecdotes is that as the number of them increases, the chances of their not being a related cause for them goes down. You learn nothing at all about wether they’re not all colluding, or independenly lying for similar reasons, or mistaking something else for what they all think they’re talking about…

If you’re only collecting anecdotes that agree with you, of course, the problem is worse by tenfold. I bet I could easily get ten thousand highly plausible anecdotes of people who flipped a coin and it came up heads. From all those anecdotes, can I conclude that all flipped coins always come up heads? Or even that they usually do?

Then they’re right. Of course, unless you go dig up a little more evidence than the mere relation of the anecdote, you’re not going to know that, are you?

There is a difference between proving something beyond a reasonable doubt, and proving it beyond all doubt, as you noted. Something can be extremely plausible to a reasonable person, and entirely false. Absent a legal context, I’d expect that ‘proof’ means ‘proof beyond all doubt’. Of course, I’ve taken more classes in logic than law, too.

The purpose of my example was to demonstrate that wether it’s one person or ten thousand claiming they talk to God, it’s not convincing evidence by itself, and it’s certainly not proof. Anecdotes have only rhetorical value; even in quantity; they prove nothing. As the claim becomes less implausible, the rhetorical argument becomes more believable, but that’s because it has less opposition, not because it’s a compelling proof on its own.

I will concede that there are situations where rhetorical arguments are effective; cases where there isn’t proof to the contrary being the most obvious example. Still though, absent independent confirmation that at least some of the anecdotes are true, they are not proof in any real form.

42

Your experience is proof to you. To everybody else, it’s hearsay. See the difference? To other people, the documentation makes all the difference.

43

An anecdote is not a datum. It is merely an an assertion that a datum exists. No amount of assertion can confirm that a datum actually exists. Multiple assertions that data exist do not amount to data in and of themselves. We have multiple examples of people claiming to have been abducted by flying saucers and we still lack a shred of evidence that any such thing has actually happened.

44

A small town was just wrapping up a long project of installing a town sewer system. There was a town meeting about the final details of the project and about having a party to commemorate the accomplishment. The mayor was proclaiming the project’s great successes to the city council when townspeople started streaming into the meeting. The first one said “My house gets flooded with raw sewage every time it rains now’” The next one said “There has been over an inch of raw sewage in my front yard 4 times this month.” This went on one after another until over half of the homes in the town had registered their verbal disgust with the project and the problems that they had personally experienced because of it. It took hours to listen to the complaints because there were several hundred.

The city council looked very worried and looked to the Mayor to say something to address what he had heard. The wizened old mayor stood up and said “My dear people, the plural of anecdote is not data! Please join us at our celebration next Saturday”

45

I can well imagine a good mayor might send out a few contractors to assess the problem before the town shelled out any money. Personal anecdotes are a often a fine inducement to start an investigation, but, depending on your goals, they might not be sufficient by themselves.

46

Upon reflection and the helpful contribution of others, I see that there is certainly a nuance (that being that the word ‘anecdote’ connotes a certain unreliability that accompanies a recalled (as opposed to recorded) event) to the statement “the plural of anecdote is not proof” that makes it valid, at least in a scientific context. Additionally, I think that in most instances it is safe to say that the statement “the plural of anecdote is not data” also relies on the same nuance to stand as generally true.

However, moving from this self imposed hijack (can it even be a hijack if it’s in the OP) to my real beef, is anyone still willing to say that “personal observations/experience s are not evidence?”

Again, what seems to be the prevailing idea of what “evidence” is really got me going, and it came after seeing it several times in a row in recent threads: for example, someone would detail a laundry list of circumstances that lead to a conclusion and another would reply, “where’s your evidence?” Whaa? Or, one might say, “it seems as though in areas heavily populated by Mexican immigrants, there are a lot of bi-lingual signs.” to which another would reply, “your observation is not evidence.” Whaa, whaa?

47

The problem, I think, lies in the proliferation of persons on anonymous fora to generate anecdotes to *make themselves an expert * on a given topic. It’s one thing to join a discussion and describe your experience of the topic - quite often that’s actually even data (probably not the best source of data - or the most objectively reliable, but data all the same). It’s a beast of another color to essentially say “It happened thusly in my case, therefore that must be the way it always is” or to say “I had an experience once relating to the topic and therefore I am an expert on all facets of this topic! Fear my knowledge and bow before my superiority!” That is the attitude that lies at the root of the “anecdotes aren’t data” rubric.

Anecdotes have validity in most discussions - they can even be considered data in a lot of discussions. (Pause for the anecdotes are never data crowd to gather their ire). This is, however, comes with the same burden as any other data - if it cannot be verified (or repeated), then its value is negligible.

Example: a discussion of raku pottery techniques is raging in GQ (hey, much weirder things have happened in GQ than raku pottery). Several years ago, I made some pottery using the raku technique - so I offer up what I learned (say it was that in my experience atmospheric conditions will affect the glazing result on a raku-fired pot - which is true. If it rains like a bastard, then the moisture will often seep into the firing if you do an outdoor firing and give you color results that weren’t what you were expecting.) It’s anecdotal, but it’s evidence. It wouldn’t be hard to verify I took some pottery classes as an undergrad and what the curricula of those classes were. Now, if I set myself up as an expert based on one class taken as a relaxation class my senior year in college - that would make me the kind of poster the “anecdote isn’t data” crowd abhor.

And the plural of “verifiable anecdote” is actually data - that’s the keystone of science. The problem lies in the “verifiable” bit, of course.

48

Oh, this has pissed me off for ages.

There was a thread in GQ about what prostitutes use when they have their periods. I posted that I had watched a documentary about a Melbourne brothel, in which one of the girls was shown purchasing sponges, and she said that this was what she and all the other girls used. Now, I didn’t say ALL prostitutes used sponges, I said there was evidence that at least one did.

I was then told by someone to go and find a primary source, because anectdotes aren’t data!

Hello, a contemporaneous video interview with a genuine real-life prostitute is a primary source!

Anyway- while it is not good practice to base everything on anectdotal evidence, or to generalise from it, it is also not a great idea to dismiss it completely.

49

the problem with using anecdotes as proof is that they are not rigorous. I understand that was said before, but not applied porperly, I think. I cna offer an anecdote to illustrate, but not prove, my statement.

My ex-brother-in-law was unemployed for several months. He told me that he had been turned down by three different prospective employers, because he was a German-American and the interviewers were Jewish. You don’t have to believe my story at all, but it illustrates the way that an anecdote only offeres information that defends a particular point of view. My xBiL was supressing information that supports other explanations. People do this when they relate anecdotes. If you didn’t supress some information, you wouldn’t have a story.

50

I said it before in this thread, and I’ll say it again here:

There now. That should be perfectly clear to all involved. :smiley:

51

The problem with your ancedote is not the documentary itself, but that we would only have your word for it that the documentary exists at all. A link to the documentary itself would constitute primary evidence, but the per se claim you saw a documentary amounts to nothing but an unverified assertion that a datum exsts. It’s not a datum in and of itself.

52

But irishgirl has been posting here for ages. To the best of my knowledge, she has not been called out for being an untruthful poster. Therefore, her claim that she has seen such a documentary carries validity with me. And, since I have some respect for her intellegence, her claim that she remembers about sponges and prostitutes is worth considering. Also, because I know that sponges can be an excellent way to keep sex from getting messy at that time of the month, her assertion is further believable.

Now if IJUSTJOINED with his guest membership posts and says that he has seen a documentary where ALL protitutes take time off at that time of the month, I’ll give it less weight. I don’t know him, nor do I know if he retains information with accuracy. The claim of ALL lacks validitity because it is absolute, as well as because what I know of prostitution and menstrual cycles, while some prostitutes could afford to take a week off each month, others are in dire straights and would find some easily accessible method to allow them to continue to work.

53

Subjective evaluations of the credibility of the storyteller still do not amount to verification that the storyteller’s alleged datum actually exists. “I believe her” is not evidence.

54

I disagree. Its evidence. In a court it would be circumstantial evidence, but that doesn’t mean it isn’t a data point. Hell, even when I deal with DATA, real numbers and measurements, its only as good as the measurement system.

(One of my personal pet peeves is that anyone who thinks that just because its formatted as what is generally accepted as data its good. A lot of data is junk, collected by flawed methods or by miscalibrated machinery. Or its used wrong - applied to a different population - Hey, its a number, its data!)

The entire field of Anthropology is based on such ancedotal observations and stories. So is History. When you are analyzing the letters of Civil War soldiers, you have no idea how much TRUTH is there, but enough like experiences draw a picture.

55

The documentary might be a put on, the girl might not be a prostitute, etc. Is the only unimpeachable data that which we produce ourselves (but cannot share with others since then it would be anecdotal)?

56

It’s not a data point, it’s an assertion of a data point. The alleged documentary itself cannot be examined or even verified to exists. I am talking about scientific standards of evidence here and from a scientific standpoint, assertions are not data.

I agree insofar as anyone actually thinks that. Scientific standards don’t work that way, though.

No it isn’t. Anthropology involves all kinds of hard data. Anecdotes, in themselves, would never be taken as proof of anything in the field, although they could be investigated to see if they can yield hard data.

Any historian would tell you that the field is full of uncertainty, that much of what is accepted is accepted provisionally, and that lots of our assumptions about history cannot be proven to a scientific certainty.

57

When did I say I would accept the documentary at face value? You’re actually proving my point. As long as the alleged data cannot be verified and examined, it’s empirically unsound to accept the allegation itself as a datum.

I don’t understand what you’re asking. What do you mean by data “we produce ourselves?”

58

I wasn’t really arguing your point, that was my point. When does ANY data pass muster? When you yourself verify it and examine it. So all data is worthless unless we measure and compile it ourselves?

Sorry, “compile” ourselves.

It seems like no matter what the data or the source, it will always be suspect. The only data that I can view with certainty is what I measure and compile on my own, yet as soon as I share that, it’s anecdotal and cannot be viewed as accurate by anybody else. It’s up to them to verify and examine it, so why bother with ever looking at anybody else’s data in the first place?

I know I’m going a little far with this, but every scientific observation is really just an anecdote. I put a burner under a glass of water, measure it’s temperature, record the readings of the thermometer. I post the results on a message board and now my readings are a worthless anecdote, moments ago they were a scientific experiment.

59

Ah, see, I was unaware that the SDMB was a forum that adhered to SCIENTIFIC standards of data on questions like “what does a prostitute do during her period.” Since its unlikely that’s been scientifically studied, the best we can do is speculation and ancedotal data. Personally, I prefer ancedotal data (like “I saw this documentary”) than speculation.

60

I’m saying that the only data which is valid is that which can be verified and examined, yes. It doesn’t necessarily have to be personally examined to be accepted but the data has to at least be available for verification. Once again an assertion that eveidence exists is not evidence in and of itself.

But these experiements are all repeatable and verifiable for anyone so inclined. You don’t have to take anyone’s word for it and that’s the difference.