My Nurse's World: LESSON 7

LESSON 7 INTRODUCTION TO BIO-STATISTICS

Biostatistics is a scientific discipline that is responsible for the application of statistical analysis to different issues related to biology. It can be said that biostatistics is an area or a specialization of statistics, the science dedicated to the quantitative study of all kinds of variables.

1. MEASURING SCALES

Different scales are used to measure variables.

Nominal Scale.

Its purpose is to identify subjects / objects within a distribution, so we can only establish the relationships of equality / inequality between the subjects / objects of a distribution. An example is the name of the cities: we can only differentiate them according to this scale. The number of soccer or basketball players gives us the same information: it only serves to identify and differentiate them from other players, we can not establish any kind of order or gradation based on this number.

Ordinal Scale.

This type of scale is intended to order the subjects / objects of a distribution according to some characteristic. It should be noted that the distance between their units is not uniform. In this way, we can say that A is above B, but not that it is double or that it is half one another. An example is the order of arrival in a race. In addition to the attribute of equality / inequality, in this scale we can add the ordering of its components.

Range of Interval.

In this scale the distance between the units of measurement is uniform, so that we can say that D is twice as much as A, for example. Therefore, it allows to perform mathematical operations, such as addition, subtraction, multiplication or division. The zero is arbitrary, it does not indicate the absence of attribute. The scale of time we use can be used as an example: zero is arbitrary, placed at the birth of Christ, or the scale to measure temperature in degrees centigrade, where zero is also relative.

Scale of Ratio.

Similar to the interval, with the only difference that the zero on this scale does indicate the absence of attribute, is absolute zero. As an example we can indicate the height in centimeters, or the weight in grams. In both cases 4 is double that 2 (2 + 2 = 4), or 4 is half that 8, for example, because the distance between its units of measurement is uniform.

1. TYPES OF VARIABLES.

QUALITATIVE

Nominales. There is no difference of importance.

- Dichotomics: because it has 2 levels or categories (axis: man, woman). Everything that is answered with yes or no is dichotomous or only with two variables like the sex, sick or not...

- Policotómicas: More than 2 categories. (Example: single, widow, married, separated), race, religion...

Orders: They establish an order, a hierarchy. Satisfaction at work:

Very satisfied. Satisfied. Little satisfied. Nothing satisfied.

QUANTITATIVES

They are the ones that can be measured in numerical terms. They are used on interval and reason scales.

- Discrete: They can only take a finite number of values. The measuring unit can not be fractionated.

Example: No. of children: 1, 2, 3, 4 ...CIGARETTES...

- Continuous: Those that can be worth any number within a range. The unit of measurement can be subdivided in infinite form.

· A discrete variable we can not convert it to continuous, in reverse yes.

The categories must be constructed with completeness (that all subjects can be classified at some point in the scale) and exclusivity (they can only be included in a category)

2. VARIABLES: REPRESENTATION OF DATA.

Frequency tables: Image of data showing frequencies in columns and categories of variables in rows.

Requirements:

- They are self explanatory.

- They are simple and easy to understand.

- They have title, brief and clear.

- Indicate place, date and source of information.

- It includes the units of measure in each head.

- Indicate the basis of the relative measures.

- Make the abbreviations explicit.

Relative frequency, is a value between 0-1, is studied by dividing the absolute frequency between the total number of the sample. For example doctors (658) among the 2343 professionals gives the percentage. Hi = fi / n

The sum of fi = n or the sum of hi = 1; the sum of the percentages must give 100%

3. CONTINUOUS VARIABLES: REPRESENTATION OF DATA IN A FREQUENCY TABLE.

4. GRAPHIC REPRESENTATION.

· Fast way of communicating numerical information (frequencies), visually, with bar diagrams

· They are the images of ideas (bars, histograms, sectors ...).

· Increase written information, provide visual guidance.

· They do not replace the text, but as support.

· Basic rules:

Visually clear

Clearly described in footnote and in text.

Graph the conclusions of the study.

Avoid confusing, not overloaded graphics.

MOST FREQUENT LOCAL REACTIONS.

Bar diagram: used to measure a qualitative variable, nominal, and especially the policotómicas.

There are variants of the bar diagram known as pictograms that are used to represent qualitative variables, it differs from the diagram because the bars are replaced by icons or images that represent what we are studying.

HISTOGRAMS AND FREQUENCY POLYGONS

Histogram: Same as above, the difference is that it is used for continuous variables. If the interval amplitude is the same, we will raise columns attached to the height of the corresponding frequency. If the amplitude of the interval is different, the area of the column will be proportional to the frequency represented.

5. GRAPHICS

TRUNK AND LEAF GRAPHICS.

Ways of expressing quantitative variables, particularly continuous.

GRAPHICS OF SECTORS: They are used to work with QUALITATIVE variables. Preferably for variables with few categories such as dichotomous or maximum 3-4 categories.