Quantitative methods emphasize objective measurement and statistical, mathematical or numerical analysis of data collected through different polls and questionnaires, and surveys, or by manipulating pre-existing statistical data using computational techniques.
It involves an intentional introduction, removal or change of the variable. We try to change the existing outcome or disease by making a change to exposure or a factor under study. The variable to be manipulated is the independent variable.
The findings of experimental groups need to be compared with the controlled group. We can have either one control group or more than 1 control group. The minimum of one control group is needed in an experimental study.
It is important because subjects have an equal chance to be placed in a control group or experimental group. It reduces the selection bias.
There should be a valid comparison. Compare apples to apples, don’t compare apples to oranges. It is not about fruit, it is about comparisons.
It is a way to divide the collection of participants into two or more groups to increase our confidence that the groups do not differ systematically. It is a mechanical method and it is automatic.
An examination that measures the dependent variable of an experiment before the treatment.
An examination that measures the dependent variable of an experiment after the treatment.
The participants who receive the treatment in experimental research.
The participants who do not receive the treatment in experimental research.
You are familiar with other terms like random assignment, dependent variables, etc.
Blinding is a procedure in which one or more parties in a trial are kept unaware of which treatment was received. Blinding is an important aspect of any trial done to avoid and prevent conscious or unconscious bias in the design. Blinding is done to prevent bias.
Types of experimental research
There are three types of pre-experimental study
Note: Examples are not mandatory, given here for better understanding.
This is also called the one-group posttest-only design. There is only one group, a treatment, and a posttest. There is no random assignment.
You take a group of forty newly hired wait staff and give all a 2-hour training session in which you instruct them to introduce themselves to customers by the first name and to check on the customers, asking, “Is everything fine?” 8 to 10 minutes after delivering the food (treatment). The participants begin employment, and you record the amount in tips for all for one month (posttest score).
There is one group, pretest, treatment, and posttest. It lacks a control group and random assignment. Measure the dependent variable before and after the treatment. It is an improved method.
You take a group of forty newly hired wait staff and give all a 2-hour training session. You instruct the staff members to follow a script in which they are not to introduce themselves by the first name and not to return during the meal to check on the customers. All begin employment, and you record the amount in tips for all for one month (pretest score). Next, you “retrain” all 40 participants and instruct them henceforth to introduce themselves to customers by the first name and to check on the customers, asking, “Is everything fine?” 8 to 10 minutes after delivering the food (treatment). Over the second month, you record the number of tips for both groups (posttest score).
Also called the posttest-only nonequivalent group design. There are two groups, a posttest, and treatment. It lacks a random assignment and a pretest.
You give forty newly hired wait staff an identical 2-hour training session and instruct all to follow a script in which servers are not to introduce themselves by the first name and but to return during the meal to check on the customers. They can choose one of the two restaurants at which to work, as long as each restaurant has twenty people. All begin employment. After one month, you “retrain” the twenty participants at restaurant 1 (experimental group) and instruct them henceforth to introduce themselves to customers by the first name and to check on the customers, asking, “Is everything fine?” 8 to 10 minutes after delivering the food (treatment). The group at restaurant 2 (control group) is “retrained” to continue without an introduction or checking during the meal. Over the second month, you record the number of tips for both groups (posttest score).
Limited control over the independent variable. Explain the Causal relationship more certain than pre-experimental studies. Explain the causal relationships in situations in which classical design is difficult. Some have randomization but lack pretest.
There is Pretest/posttest. It requires the collection of data on the study’s participants’ level of performance before the intervention took place (pre-) & then the collection of the same data after the intervention took place (post-). Allows making inferences on the effect of intervention by looking at the difference in the pre-test & post-test results.
With comparison (placebo is given to the control group)
Without comparison (placebo is not given to the control group)
Identical to the static group comparison. No pretest and random assignment in the static group. Here RANDOM ASSIGNMENT. It has all parts of the classical design except a pretest.
We can introduce treatment and also remove it. The dependent variable goes up after treatment. And it goes down when treatment is removed. This is evidence of the effect of treatment.
There are three parts i.e.; Pretest-treatment-posttest. When there is the removal of the treatment-posttest. There will be a change in the effect of treatment. When we restore the treatment-posttest, then again there will be a change in the effect of treatment. In this way, we can check the effect of treatment.
A comparison (control) group does not receive treatment. E.g., if a treatment group starts below the control Group & ends up above after treatment, a stronger influence of a treatment effect exists.
Group of people with similar characteristics within a defined period.
THE EFFECT OF TOBACCO USE ON ADULTS WHO WORK FOR PAY.
This design can limit the confounding variables
This design is cheaper than RCT.
No Control group, no randomization. A series of observations on a dependent variable over time. Interrupted by the introduction of an intervention. The time series should show an “effect” at the time of the interruption.
An experimental plan with several repeated pretests, posttests, and treatments for one group often over some time. An equivalent time-series design is a one-group design similar to the interrupted time-series design. It extends over some time, but instead of a single treatment, the equivalent time-series design has the same treatment multiple times.
The propensity score is the probability of receiving treatment. This is a statistical matching technique. This is used for the effect of treatment and co-variables. It explains the causal relationship in observational studies. This is the estimation method.
|Most powerful design to establish a causal relationship between dependent and independent variables.||Studies conducted in hospitals and communities, difficult to control extraneous variables.|
|Provides better results||Several issues such as ethical reasons can hinder the experiment|
|Allow the researcher to have precise control over the variable.||Participants can be influenced by the environment|
|Provides stronger evidence||Time-consuming|
Experimental studies translate results of basic scientific research into better ways to prevent, diagnose or treat disease.
Basic and Clinical Biostatistics by Beth Dawson and Robert G. Trapp
Pearson’s new International Edition, Social Research Method: Qualitative and Quantitative Approaches.
Introduction to statistical methods for clinical trials by Thomas D. Cook and David L. DeMets.
The concept of Blinding in clinical trials-EUPATI.
About the Neonatal Intensive Care Unit The Neonatal Intensive Care Unit (NICUs) has compound equipment, staff and devices for the… Read More