|
Year |
Male (at birth) |
Female (at birth) |
Male (at age 65) |
Female (at age 65) |
|
1940 |
61.4 |
65.7 |
11.9 |
13.4 |
|
1945 |
62.9 |
68.4 |
12.6 |
14.4 |
|
1950 |
65.6 |
71.1 |
12.8 |
15.1 |
|
1955 |
66.7 |
72.8 |
13.1 |
15.6 |
|
1960 |
66.7 |
73.2 |
12.9 |
15.9 |
|
1965 |
66.8 |
73.8 |
12.9 |
16.3 |
|
1970 |
67.1 |
74.9 |
13.1 |
17.1 |
|
1975 |
68.7 |
76.6 |
13.7 |
18.0 |
|
1980 |
69.9 |
77.5 |
14.0 |
18.4 |
|
1985 |
71.1 |
78.2 |
14.4 |
18.6 |
|
1990 |
71.8 |
78.8 |
15.0 |
19.0 |
|
1995 |
72.2 |
79.1 |
15.3 |
19.1 |
The Latin word nomine helps to describe the nominal data type. Nomine means name, and that is precisely what nominal data do; they name objects. What the nominal data type does NOT do is important too. Nominal data have NO MATHEMATICAL PROPERTIES. It is important that you understand that last point. Some examples may help to clear up this concept. Sport teams players usually wear uniforms with numbers on them that are strictly nominal. So, a football player can be identified by the number on his uniform (often, the jersey number also identifies his position too). Another example is a code number assigned to you for the posting of grades. Social Security numbers are still another example, they simply identify people with regard to their retirement benefits. Since the passing of the Buckley Amendment, the public posting of students grades has been illegal. However, posting student grades with a private coding system is legal, and private. Your code identifies you and your grade, and if you do not tell others, is known only to you. Obviously, you cannot perform any statistics on nominal data.
Another Latin verb ordinare means to put items in order. That describes the ordinal data type well. Ordinal data have the mathematical property of indicating order or place. For instance, knowing that a horse finished second is an ordinal datum. Notice that just knowing that second place finish does not tell you anything about how close or how distant that place was. You may perform basic statistical tests on ordinal data (medians, %centiles, and rank correlations).
Interval data do provide information about the distance between data points. A scale is a very common way that social scientists provide interval data. Good physical examples are the common temperature scales, centigrade and fahrenheit. All interval data assumes equal divisions between data points. Note how the examples above do that. But, IQ tests and many other scales lose that property at their extremes. For instance, arguing about who has a greater level of intelligence between individuals with extremely high IQ scores (i.e, 160+) is not useful. With interval data you may perform a wide variety of statistical tests including means, deviations, correlations, and tests of significance.
Ratio data have equal distances between units and an absolute zero point that is not arbitrary. The real zero point is the main characteristic that distinguishes them from interval scales (another distinction is that many interval scales' assumption of equal distances between units is not true).The typical examples are geometric distances and the Kelvin scale (Ko) of temperature. The former is a ratio scale because you can measure a distance of 0 meters (i.e., the object did not move any distance). The Kelvin scale is probably the best example of all because there really is no temperature below 0oK (all molecular motion ceases).
Probably the easiest way to understand what an independent variable (IV) is to think of it as the reason for conducting your research project. Experimenters select their IVs, hence their name. IVs are also known as levels or treatments.
Probably the easiest way to understand what a dependent variable (DV) is to think of it as the way of measuring the effects of the IV. Whenever you think of a DV, think of a way of measuring the IV in your research project. DVs depend on the IV, picking a DV is akin to an art, and the more you do it, the better you will get at it.
Every other possible variable is an extraneous variable. In experimental design, experimenters strive to reduce or measure the effects of extraneous variables. Later in the Methods sequence you will learn how to do so.