Foundation of Statistics STA10003- WEEK 01 NoTEs
Statistics is used as an overarching term that includes the procedures and techniques that are used to organise, summarise and interpret information collected. Within statistics further terminology is used to represent things as populations, samples and variables. Terminology 01. Population (N)- it is all of the things we are interested in. For example a population of people, household, trains ... etc. It is the parameter that describes the entire population and it is difficult to measure. 02. Sample (n)- a subset of population which is representative of the population that we refer to. 03. Variable- it is an overarching label of the element/ feature we want to measure. 04. Sampling error- it is the natural error existing between the sample statistics and the population parameter. Descriptive Statistics It organises raw information into manageable information numerical/ graphical basically summarizing the data. Inferential Statistics Technique that allows us to use sample statistics to generalize and make conclusions about the population. Descriptive research When we need to find something out we obtain descriptive information. For instance, the age of the students studying a particular unit. According to this example we collect information in order to know something therefore it is a descriptive research. Correlational research We use correlational research when we want to explore a relationship between two or more variables. Comparative research We use comparative research when we want to compare 2 or more groups on the same measure. Research allows us to study the cause-and-effect explanation. It helps us to see whether the independent/predictor/explanatory variable (IV) causes a change on the dependant/outcome/response variable (DV). This is a typical experimental method. When it comes to the non-experimental methods there is no cause-and-effect explanation. Both variables are simple observed and measured. Correlation research, non-equivalent groups, and pre-post studies are examples of non-experimental methods. For instance, annual income of females and males is an example of non-experimental research because here the gender cannot be allocated. We simply observe and take the measurements. In non-experimental research the IV is called a 'quasi-independent variable'. Discrete variable It has separate categories with no intermediate values in between. It is a countable variable and the data is discreate. Example: Gender. Continuous variable Here we have infinite number of values that fall between two observed meausured values. This is divisible into an infinite number of fractional parts. Example: Age. 1
Nominal Used to label the group. Here we use a label which has no meaning. The order can be changed. Ordinal Used to label. Here we have an order. Interval Here the numbers have a meaning, order and the intervals between the numbers are equal. The numbers are continuous. Here the value also the value zero has a meaning. Example: Temperature in C or F Ratio The numbers have an order and the intervals between the numbers are equal. Here the value zero means a complete absence of something. Example: Numbers of correct answers in a test. Scales of measurement. Categorical: Nominal