Inductive Significance

Date - July 3, 2009

Time - 18:18:18

Place - Couch underneath my bunk bed at home

Listening to - Palestrina - Lamentations of Jeremiah the Prophet: Book IV Feria V in coena domini Lectio 2 by Giovanni Pierluigi

I like numbers and data. I like having specific details like date, time, and place. I like statistics.

I really enjoyed my nursing statistics class this past semester (I think I was the only one who liked it) and was especially intrigued when I began to apply what I was learning to life. Unfortunately, I've never been much of a theorist. I admire those who have inductive reasoning skills. They can take small details and develop an overshadowing theory or principle. Deductive reasoning is more my thing. I can take large principles and break them down into smaller bites. However, statistics class this past semester was a different story. I began to theorize. I was able to take the small details I was learning and piece together principles for my life. I'm still not much of a theorizer but here is an attempt:

A few necessary terms:

Research hypothesis - statement that says that a relationship exists between two or more variables and differences observed are caused by a specific intervention.

Null hypothesis - statement that says there is no relationship between to variables. This is accepted as true in the absence of other info until research proves otherwise.

Significant - any difference between attitudes of two groups that is due to some systematic influence (i.e. not by chance)

Significance level - the risk associated with not being 100% confident that what you observe in an experiment is due to the manipulation you imposed

Statistical significance - the degree of risk that you are willing to take to reject the null hypothesis when it is actually true

An example of a null hypothesis: There is no correlation between the rise of gas prices and the rise of ocean levels.

Researchers have to decide whether to accept or reject the null hypothesis. They make this decision by estimating the probability that the difference observed could be obtained by chance alone. If this probability is less than the predetermined significance level, the data is deemed to be statistically significant and the null hypothesis is rejected.

So, whether the null hypothesis is accepted or rejected is all based on a predetermined number.

Here is a chart depicting what might happen:

Significance is usually set at 0.05 or p<.05. That means there is a 5% chance of rejecting a true null hypothesis (Type I error)

Why not set it the significance level at 0.0001? There would certainly be less chance of error! But there would also be a much greater chance of missing something incredibly significant. That would be a Type II error - there really was a correlation between the rise of gas prices and the rise of ocean levels, but your significance level was so tight that you didn't notice the relationship. So you accepted the null when it was actually false.

My theory:

I think this principle can be applied to human relationships as well. If our fear of error is so great that we set an incredibly small significance level, we're liable to miss something earthshaking. But if our significance level is too big, than we are more likely to be making errors.

You'll have to do the applying to your own individual situation, but I've found it to lead to some intriguing observations in my own life. It helps if you figure out what the null hypothesis would be in your specific case and then fill in the chart with different scenarios depending on whether the null is true or false.

Limitations.

1. This is life we're talking about. We can't just run a quick statistical test to find out whether the null hypothesis is true or false. Unfortunately, (or fortunately) people don't always compute down to data. Therefore, this theory is good for thinking a situation through, but it won't give you hard facts until time passes and you discover by trial and error whether the null was true or false.

2. Again, humans are not numbers. We have these feelings called emotions that tend to affect our decisions. I'm still trying to figure out what part emotions play in my theory. Do emotions render mathematical theories applied to humans as futile?

3. My biggest conundrum is this: what role does God and the Holy Spirit play? Does He shatter all theories? Or does He lead us to choosing the perfect significance level for our situation?