This is a five-section course as part of a two-course sequence in Research Methods in Psychology. This course deals with descriptive methods and the second course deals with experimental methods.

Loading...

來自 Georgia Institute of Technology 的課程

Descriptive Research Methods in Psychology

2 個評分

This is a five-section course as part of a two-course sequence in Research Methods in Psychology. This course deals with descriptive methods and the second course deals with experimental methods.

從本節課中

Module 1: Introduction

- Dr. Anderson D. SmithRegents’ Professor Emeritus

School of Psychology

Anderson Smith again, and we're going to continue that discussion

of Descriptive Research Methods in Psychology.

And now, I want to just give you an introduction to descriptive methods,

and how descriptive methods are really different from experimental methods

that we talk about in another course.

Now, psychological research, as I said,

can be divided and descriptive and experimental research.

And descriptive research that we're talking about in this course

is used to describe characteristic of population of phenomenon being studied.

In other words, we just want to describe

what are the behavioral characteristics of mind and behavior.

We cannot answer questions about how or why the characteristics occurred.

That requires experimental techniques,

but it's very important to have these good adequate descriptions.

We often collect just actual frequency information.

How often does something occur, and that you can do with descriptive techniques.

We also can examine the relationships among variables using correlations,

whether or not two measures are related to each other

and how they relate to each other and how much.

And we don't have to manipulate the environment in any way,

we just want to study what's actually there.

And that's what descriptive techniques do.

Now, with experimental techniques, we do use- we do study phenomena

and determine causal relationships among variables.

We use manipulation of a variable, and with controlling everything else,

and then look at whether or not that affects the behavior we're interested in.

So we had to manipulate one variable at a time

and control all the other potential variables

that could affect the measurement of the phenomenon, the behavior that we're looking at.

So we test hypothesis to reach conclusions, and then hypothesis

is a prediction about what the relationship is going to be,

what the causal relationship between variables actually exists.

Now, there are different kinds of descriptive methods and we're going to talk about those.

There are case studies, there are observational studies,

different kinds of observational studies

where I simply observe behaviors in their natural habitat.

There are surveys and polls, and then there are correlational studies.

And we're going to look at all of those today.

First I want to just give you a little tiny introduction to descriptive statistics.

How we go about talking about the observations that we make.

First there are two kinds, the measures of central tendency so,

what is the average of the behavior we're looking at.

And there are three different kinds of measures of central tendency,

the mean, the median, and the mode.

And then the measures of variability have dispersed other measures,

and there are three measures are that, the range, the variance and the standard deviation.

The central tendency.

The mean is just the average.

You sum up the scores.

And then you divide it by the number of subjects you have and you get the average score.

And that's all the mean really is.

The median is the score that divides half the scores from the other half of the scores,

so it's the middle of the distribution.

And sometimes a measure of central tendency is better to use the median than the mean.

For example, if I measured how fast people react to something,

and I have most of the reaction times sort of centered around a few numbers

but there are a couple of people who have very, very long reaction times,

than giving you the mean would be misleading because of these outliers.

So, another measure of central tendency would be the median.

What score falls in the middle of all the scores that I measure.

And then finally, we have the mode which simply is the score with the highest frequency.

What is the most frequent score that I get,

measure that I get in the distribution of measures.

Now, let's take an example.

Here we have some scores one, three, eight, eight 10, five different scores.

The mean would be the sum of those scores,

and then divided by the number of scores that I have.

So that would be 30 which is the sum divided by five.

So the mean would be six.

The median to be the one in the middle

and the one in the middle as you can see is an eight.

So the median is eight and the mode is the most frequent.

And I have two eights.

So the mode would be eight.

So, there are different measures of central tendency.

There are also, this is another example that I have sort

of what is the proportion of people that have different family incomes.

And you can see the median, and this case is different from

the mode and the mean, the mean would be about 74 k.

The mode will be 50 k right in the middle,

and the median the score right in the middle of the distribution.

I said mode was in the middle, mode is the most frequent,

and median is the one that's in the middle which is 61 K.

So, different measures of central tendency depend upon

what the distribution looks like and how it's skewed.

In variability we have the range,

which is simply the difference between the lowest score and the highest score.

What is the total distribution differences.

The variance which is the average squared difference on the scores and the means.

You take the mean and how different the score is from the mean,

and then square that and that gives you the variance

and then simply the standard deviation then.

And this is an example of that where you have the sum of the scores.

Subtracting the means and squaring it,

dividing by and minus one that gives you the variance.

And then the standard deviation is simply the square root of the variance,

so it's the square root of that.

And standard deviations are important because if you have a normal distribution,

it tells us where the score falls in the distribution.

So, let's look at another example of variability, and also central tendency,

using this distribution of scores from five to nine.

The mean is the sum of the scores which is 140 divided by 20 which is seven.

The median is the one in the middle which is seven.

The mode is six in this case, the one that gives the highest number of scores.

And, as you can see, there are six sixes or six is the mode.

The range is the highest score minus the lowest score which is four.

The variance is the squared difference between the mean in each score,

and then forcing my computer [inaudible] , does squares by showing upward too.

But, you know, that's two with a little two up above it which would be four.

And then you add those together and divided by n minus one

and you get the variance which in this case is 1.58.

And then the standard deviation is simply the square root of the variance which is 1.26.

So, in this case the mean and the median of seven, and the mode is 6,

the variance is 1.58, and the standard deviations is 1.26.

All of these are just examples of measures of central tendency

and variability that we need to have in order to understand descriptive statistics.

Thank you.