BSTAT analytic goals

analytic goals

directed toward finding out from the data one or more of the following attributes or characteristics of the group being studied o refer to the specific objectives or questions that a statistical analysis aims to answer or achieve.

analytic goals

refer to the specific objectives or questions that a statistical analysis aims to answer or achieve, Without these specific objectives, you're essentially wandering aimlessly through your data, which is inefficient and unlikely to yield actionable insights.

analytic goals

transform a raw collection of numbers into a purposeful investigation. They ensure that your statistical work is not just about crunching numbers, but about extracting meaningful, actionable insights that solve problems, answer critical questions, or support informed decision-making. Without clear analytic goals, statistical analysis risks being a complex exercise with no practical outcome.

Central tendency

Variance in the Group

Difference within the Group/Between Groups

Relationships within the Group

prediction

Different Types of Analytic Goals

central tendency

To find the typical or central value of a dataset.

variance of the group

To understand the spread or dispersion of data points around the central tendency.

Difference within the Group

To compare characteristics, performance, or outcomes between different subgroups or across different conditions/ times

relationships within the groups

To identify and quantify associations, correlations, or causal links between different variables within a dataset.

prediction

To build models that forecast future outcomes, trends, or values based on historical data and identified relationships.

time factor

Reduced cost

large size of many populations

Destructive nature of some studies.

REASONS FOR SAMPLING

time factor

sample may provide you with the needed information quickly.

time factor

For example, you are a doctor and disease has broken out in the area of your jurisdiction, the disease Is contagious and it is killing within hours, nobody knows what it is. You are required to conduct a quick test to save the situation. If you try the census of those affected, they will be long dead before you arrive with your results.

reduced cost

it is obviously less costly to obtain data for a selected subset, rather than the entire population.

The large size of many populations

in some cases the size of the population is extremely large

The large size of many populations

For example, if you conduct research studying all elementary school students in a particular country, it is difficult to reach all the students because of their large number.

destructive nature of some studies

when a doctor wants to do a blood test, he only needs to take a sample because if he drains all blood out of the patient, the patient will definitely die

sample

are an important component of research, and they are used to make inferences about a population based on a smaller, representative group

sample

are used in both quantitative and qualitative research, and they are an essential tool for making generalizations about a population.

sampling frame

is the list from which the potential respondents are drawn.

example:

registrars office

class rosters

undercoverage (omission)

overcoverage (Inclusions of Ineligible Units)

multiple listings

inaccurate information

4 GENERAL TYPES OF SAMPLING FRAME ERROR

undercoverage

This is when elements that should be in your study population are missing from your sampling frame.

undercoverage

Example: If your study population is "all college students in a university, but your sampling frame (registrars list) only includes students who have paid their tuition in full, you're undercovering students on payment plans, who might have different characteristics.

overcoverage

This is when elements that should not be in your study population are included in your sampling frame.

overcoverage

While these ineligible units won't affect the final sample identified and removed before data collection, they can increase the cost of sampling (e.g., trying lo contact people who don't qualify). If not identified, they can skew results Example: Your study population is "currently enrolled students," but your class roster (sampling frame) still includes students who have recently dropped out or graduated

multiple listings

This is when the same element appears more than once on the sampling frame. Impact: Units with this have a higher probability of being selected than those listed only once, leading to a biased sample.

inaccurate information

This refers to incorrect contact details, demographic information, or other relevant data for the units listed in the frame.

Impact: While not directly affecting selection probability, inaccurate information can lead to non-response bias you can't contact them) or misclassification of respondents.

Example: The phone numbers or email addresses on the registrar's list are outdated for many students.

ensuring representativeness

improving efficiency and cost effectiveness

enhancing generalizability

strengthen research validity

Why are sampling errors critical

population is very small

extensive resources

dont expect a very high response

Can you sample the entire population

target population

This is the entire group of individual, or events that the researcher is ultimately interested in studying and to which they want to generalize their research findings. It's the broad group that meets the criteria for the research question.

Characteristics: It's often very large, sometimes infinite, and way not be entirely accessible.

Sample

(The Smallest, Innermost Rectangle, within the Study Population)

This is the actual group of participants or observations selected from the study population to be Included in the research. Characteristics: It Is a smallest, manageable subset. The goal of sampling is to select a sample that the representative of the study population