Chapter+Five+-+Quantitative+Research+Designs

= = =Chapter Five: Quantitative Research =

The purpose of this chapter is to provide a discussion of the various research designs used in quantitative research studies. Each design should be clearly described, including procedural steps regarding sampling and data collection as well as the type of statistical analysis that would be used to determine the outcome(s) of a study using that design.

1. Overview
Quantitative research involves the analysis of numerical data. The researcher knows what he or she is looking for because they have designed the study to prove their overall hypothesis. It is important to be non-bias when looking at the data which is usually very efficient and true to the original concept. The data is usually gathered at the end of the study after all the numbers have been evaluated from surveys, questionnaires or equipment.

2.Experimental Designs
This is a collection of designs in which a researcher manipulates or influences an independent variable and controls the dependent variables to test a causal-effect relationship between them. Experimental designs are commonly used in the sciences and help to make predictions and explain phenomenos. Experimental designs are usually very consistent because a cause will always lead to an effect. Key features of experimental designs include: 1) Pretest and Posttest (checks to ensure that the groups are different before the manipulation starts), Control groups (does not receive the same manipulation because it is used to measure the effect), Randomized controlled trials (comparison between the independent variable and the dependent variable). Identifying and controlling any factors which are not part of the experiment but might impact the outcome of the experiment is key to ensuring the reliability of the outcome and conclusion.

2.1 Single subject designs
A single subject design is as the title suggests. It is a design that has only one participant. This design is appropriate to use in a rare circumstance case. For example, if you had a student with a unique situation that was different from most or all in the class, then you could conduct an experiement specifically designed to meet the needs of that student and test the results according to that participant.

3. Nonexperimental Designs
In experimental designs, the researcher sets up two comparison groups. The researcher must manipulate variables in one of the groups in order to create an independent variable. This is not the case in non-experimental designs. Non-experimental designs “involve the extension of experimental reasoning to the non-experimental situation (Punch p 218). In a sense a non-experimental design in similar to a correlational survey because it has little control over the independent variables.

3.1 Causal-comparative Designs
A causal-comparative design may be used when it is not possible for a researcher to manipulate the independent variable. In this case, the researcher can use the research of established groups. For example, if a researcher wants to know how one student in the same school, same class, and receiving the same instruction from the same teacher write significantly better than their peers. The researcher will develop a basic design issue. The research question should be developed based on the research of the previous findings. The hypothesis should also be developed based on the research question. The sample has to include at least two groups that can be compared, such as the difference in writing between boys and girls.

3.2 Correlational designs
===Essentially, correlational designs attempt to study the relationship between variables. Basically, the researcher is trying to see how certain variables affect other variables directly. For example, a reseacher might attempt to study how peer tutoring affects reading fluency and comprehension in an eighth grade reading classroom.=== Correlational designs are studies that measure relationships between variables. If the researcher wants to determine how teacher's felt about teaching culturally relevant literature, then they would measure the teacher’s attitude about teaching culturally relevant literature versus the teachers that actually teach it. When one thinks of correlational designs, it is less complicated to decipher if he thought of it in terms of cause and effect.

3.3 Survey Designs
When a teacher is creating a survey design (for a survey) for his or her students to take- before a research project is started, the teacher has to take into account, what the teacher wants to learn from the students. Whatever kind of survey questionnaire the teacher creates,  it needs to have answer that is very specific and not open-ended. Questions that are __non-specific, can leave student choosing answers that are non-conclusive.__

There are two types of survey designs: cross-sectional and longitudinal. In a cross-sectional survey design all of the data is collected at the same time. The researcher comes up with a pre-determined sample based on the research question that he/she is addressing. On the other hand, in a longitudinal design the data is collected over a period of time. Longitudinal designs can get a little more complicate due to the length of the study. __Researchers__ have to deal with issues of attrition and oftentimes their sample can change over the __course__ of the research.

For instance, in a longitudinal design a study may target the motivational level of 4th grade students when it comes to the TAKS Writing test. The data collected for the study will occur over a period of time. The data will be collected for these same students until they get in 7th grade and have to take the writing test again. The longitudinal design can get complicated if the sample of the students being studied changes. The study can be impacted by losing contact with a student if they move, or if a student is retained in a grade. Then, the researcher may have to deal with attrition (losing a person) in a study.

4. Summary
Quantitative research measures the relationship between two variables (the independent variable [cause] and the dependent variable [effect]). The third variable put into play in quantitative research is the contolled variable. This is the variable we want to control in order to remove or change effects. Designs used to collect the data include experimental, non experimental, single subject, survey, casual-comparitive, and correlational designs. Each design has a specific purpose and is dependent on the questions we want answered in our research.