Shaping my perspective

The most meaningful thing that I have learned in this course is how to be a better consumer of research studies. I now read the newspaper with an eye out for the articles that cite ‘research shows..’ and wonder where all the data is to back up those statistics floating around out there. I read the articles in the Wall St. Journal everyday and in journals for school and recognize the fundamental terms, methods and designs of research studies that are described and have a better understanding of where the data is coming from, if the conclusions make sense, and if the data can be considered valid. I am more aware of issues of validity and I see a lot of information given as fact when it is based on faulty methods. I have also seen a lot of graphs and charts that are meaningless, but look like they’re backing up some important facts!

I understand the different approaches to research and how both qualitative and quantitative methods are necessary and important to the contribution of research, and how different research designs are more appropriate to particular research problems than others. I know that I will use all of the course objectives, minus one, in my graduate work, as I am just beginning my program and I am sure that I will have many research papers. This course has taught me how to critically evaluate the scholarly works to find information which is meaningful. The only objective I won’t use is the ‘prepare a viable research proposal.’ This class has pretty much confirmed for me that I have no interest in pursuing a Ph.D. I am much more interested in applying the results of research than in designing the study and spending enormous amounts of time studying one topic. This class has been very useful though in that I look at studies much more critically. It is pretty interesting how quickly a lot of us just accept the ‘facts’ because it sounds good, it came from research, or it supports what we already think we know, without critically examining the source.

Specifically, I have used the objective pertaining to ethical research with human participants in the past when I worked in a local detention center as part of a funded study. I will continue to abide by those principles if I have the opportunity to continue with that work.

I am already using a lot of the skills for a project at work. I have to develop a presentation to present a business case for our company to jump on the facebook bandwagon. Several marketing firms have really pushed the idea and have explained why it’s important, but I have yet to see the data to back up their claims. I have seen some pretty graphs that look impressive but don’t even label an axis! Now I need to take time to read up on it in some reputable journals and actually look at the data and studies behind claims that facebook is a great way to reach students for our purposes- if there are any. So, I will find studies with data, evaluate design, methods and results to determine the validity, and then translate what I have found into a coherent, concise argument that supports our desire to facebook for business purposes!

Qualitative research is a different approach to studying a topic from quantitative research. Qualitative researchers operate under the belief that in order to truly understand their subject they need to collect data from participants who exhibit the qualities, characteristics, traits, etc. that they are interested in. The researchers believe that there are multiple explanations for a particular behavior or phenomenon and there are methods to go about understanding it.

Their methods include studying the behavior where it naturally occurs for the participants. The setting is not controlled or manipulated to block the affects of ‘unintended’ variables. The researcher is directly involved in collecting data from the particpants themselves. The researchers spend lots of time directly studying and interacting with the participants and any documentation to acquire data. Data is not numeric but descriptive and the researcher collects all data provided by the participant to try and get an understanding not of what a particular behavior is, but how and why that behavior occurs. Through this perspective all data is collected because it is not apparent at this stage what may or may not prove to be useful in this understanding. These studies often occur over a long period of time so that all relevant data may be collected.

Analysis of the data involves gathering all that has been collected and then synthesizing to develop generalizations or themes that can be developed to explain the phenomenon. An important part of teasing out themes is that the researcher tries to understand the hows and whys from the perspective of the participants, not by injecting their own interpretations and meanings to the behavior.

Another important difference between qualitative and quantitative research design is that qualititative studies evolve over time. The researcher starts with a basic idea about what behavior he is studying and what data he might collect and how, but there is flexibility assumed in that as more is learned about the behavior, adjustments will be made in the methods and data  collection to accomodate the new knowledge.

Schempp’s study illustrates these main points of qualititative research. His research involved the study of a high school pe teacher to understand how the teacher acquires knowledge to teach. The context of the study was naturalistic in that the teacher was observed on a regular basis while at work in the school. Schempp collected data directly from the teacher in the form of interviews, observations, documents and videotapes. There were also interviews of collegeagues which helped to clear up discrepancies and add information.

Data analysis was ongoing throughout the course of the study and following for one year. The researcher collected and analyzed data on an ongoing basis which allowed for recategorizing and identifying themes as they emerged with each new data. Schempp explains that he categorized and developed themes based on particular procedures and specifically that the categories were based on Shulman’s theory of a knowledge base for teachers. Schempp used this as a check and eventually used those categories into which his data fit under these circumstances. There were multiple revisions to reduce the categories such that there was no redundancy. This framework was ultimately checked against the impressions of the teacher for accuracy in reporting his perspective. Schempp used data and reported results from the perspective that the teacher offered and in the teacher’s own words, emphasizing the importance of participant perspective in qualitative research.

Lastly, Schempp explains that the ongoing data collection and regular analysis incorporating new data allowed for the evolution of the design of the study. He was able to make adjustments based on the new information and the direction that the studythen took, allowing for ongoing testing of themes that seemed to present themselves.

Is the recidivism rate of women who participate in career development workshops while incarcerated lower than the women who do not?

How each threat to internal validity  needs to be considered in this study:

History- This definitely can be a confounding variable. The study itself will need to make explicit how much time will elapse after release from prison and when the rate of recidivism is determined. In other words, do we measure the rate of recidivism after 6 months outside of prison, 1 year, 5 years? Obviously, all kinds of variables could interfere with the independent variables, (career development or not), in those spans of time.

Selection- This could be a confounding factor which would be determined by who received the career development and who did not. Was it voluntary? Did the women have to meet certain criteria to be involved? Was it random based on the date they entered the system? Obviously some of these factors could determine certain characteristics of the subjects that could be confounding variables.

Maturation- This is another factor that could influence this particular study. Emotional and social maturation could influence the rate of recidivism and become a confounding principle.

Pretesting- I imagine this could influence a group, but in my opinion, for what that’s worth, not enough to be a confounding variable. With the work I did at the detention center, we gave a pretest to the group of women taking our workshop and a group that was not included to determine attitudes and preknowledge before the class. The classwork was involved enough and the women in this facility had few enough other resources I don’t think this could have had a significant enough effect to be a confounding variable.

Instrumentation- I think this wouldn’t be an issue with this particular question. Numerical data would be collected from parole officers on women who fit the criteria- went through career development or not within a specified time frame and who returned to prison or not within a specified time frame. I don’t see where bias could occur with those numbers.

Treatment replications- If I understand correctly, the number of replications would have to equal the number of women in a group (the group that had career development for example). If that is correct, the big problem I see might be in funding the research to study that many classes.

Subject attrition- I am not sure how women could drop out of the group in a way that would be significant. Perhaps if a significant number of women quit the career development class this might be a confounding variable.

Statistical regression- This might be a variable depending on some of the same factors as selection. If women were selected based on a characteristic of never having held any paid work, for example, statistical regression might play a confounding role. (If I understand the concept correctly).

Diffusion of treatment- This could be a confounding variable in a study like this. If the group with the career development workshop and the group without are chosen from within the same prison, there is a good possibility given the tight quarters that one group could influence the other.

Experimenter effects- If the data is collected from a parole offficer(s) by an experimenter I’m not sure how much bias would enter.

Subject effects- I think this wouldn’t be an effect if we are working with data and not people specifically.

This chapter was a little mind boggling with all of the details that need to be considered and accounted for that may affect the outcome of a study. It was very informative though in presenting things to consider when designing a study or evaluating one!

Although it is tempting and sometimes seemingly straightforward enough, causation cannot be inferred from information presented in correlational or comparative designs. Just because a study has shown a relationship to exist between two or more variables, we cannot make the assumption that one has caused the other. They quite possibly “coexist” and the dependent variable could have been caused by another related, undiscovered, unexamined, or unreported variable. A variable cannot be presumed to be the cause of another in a comparative or correlational study.

A comparative study can show that there does, or does not, exist a relationship between two (or more) variables. It is easy to assume that one variable caused another when a relationship has been shown to exist between the two but we must always look behind the scenes and take apart a study to determine if there are other factors that could have come into play. It is just as possible to turn that supposed cause upside down and say the reverse effect.

In Legette’s study on perceived attribute causes, he is comparing several independent variables to the dependent variables of cause attribution. The results showed that students tended to place more weight on ability and effort in considering their success. Although they feel this is the cause of their success or failure, there could be a relationship without cause. It is also possible that out of the choices, these were the best options for the students, and not really what they felt were the actual causes. Also, success could cause the children to attribute a cause that they have control over as opposed to luck, for example, which they don’t.

I’m not clear as to why Legette chose the 2 school systems he did as they are not similar and the differences in responses can be attributed to many factors other than “city” and “county”. It is hard to see what the study is hoping to accomplish by comparing such diverse populations. There certainly can be no cause established although a relationship has been established. I am also not clear as to why Legette chose to study 3 different school levels.  Although a relationship has been established, what is he explaining? The study here seems to be comparing apples to oranges and not establishing clear patterns of relationships. Also, in choosing the variables such as gender and school level and presenting the differences in data the way he does, it seems he is implying a causal relationship, especially since his questions overtly ask children about the perceived cause of variables. It is easy to get caught up in the data and amount of information, including language used, and assume that a comparative study (or correlational) is attempting to show cause, when as seen in this case, the data strictly is useful from a comparison point of view. And in this particular case, I’m still not clear on what it its I would do with the information this particular comparison presented.

I believe there are several important ethical considerations that a researcher must abide by when conducting research with (human) participants. First, the participants must be aware of the fact they are being studied! This involves taking the step of disclosing to them the purpose and methods of the study and what their specific roles and responsibilities will be. There also needs to be disclosure as to possible benefits or detriments to being a participant. The researcher should take care to make sure the potential participant truly understands these possible outcomes and the probability that one or another might occur. The person must have time to consider all aspects before making a decision about whether to participate or not. The person must be capable of making a decision of this sort for themselves (or the parents or caregivers of the participant if he is a child). Participation should always be optional and there should be no penalty for ending participation at any stage of the study.

It is the researchers responsibility to protect the participants from physical and emotional harm. This is particularly important as there is a level of trust established in the disclosure-acceptance stage that is fragile. Not protecting the participants could cause emotional harm beyond that specific to an incidence. This includes keeping data and personal information confidential. It cannot be assumed that because a participant shares with a researcher that they are willing to share with anyone else.

The principal researcher is responsible for the handling of participants and data and for employing other researchers who will adhere to the standards of the study. The head researcher also is responsible for making sure that the standards are being followed.

Overall, there needs to be a level of respect for the participants as human beings. A researcher cannot lose sight of this fact in his or her quest for data.

Is the recidivism rate of women who participate in career development workshops while incarcerated in low level security prisons lower than women who do not?

This is a quantitative research problem. The results will hopefully be generalizable to a large population. The results will show the efficacy or not of in prison career development programs on post release behavior.  The population is former women prisoners of low level security prisons. In choosing the participants first I would look at simple random sampling. The study could use the records from all low level security women’s prisons from several states. The participants could be chosen at random from the thousands of women in the target population. It could be assumed that using a large population across several states would produce a representative sample of former women prisoners as a whole. It would be fairly easy to produce a large number of participants this way. The problem with this sample method might be the time consuming aspect of numbering all of the possible participants in order to choose a truly random sample.

Systematic sampling may be a more useful way of choosing a random sample. The computer records of the target population could again be assigned numbers with the computer choosing a random number to start drawing a pool of participants from and then continue the pattern of choosing. This would produce a truly random sample. Although there is concern with periodicity in using lists of participants with this method because of how participants may be categorized, thus leading to a uniform and not random sample, I believe that would not happen in this case because records could be sorted alphabetically first and then assigned a number.

Proportional stratified sampling would involve breaking the target population into subgroups and choosing a random sample from each of the subgroups. I believe that this is the method that I would use to address my research problem. In trying to arrive at an answer to the problem I would need to look at the recidivism rate of two groups of women to be able to compare the results. One group of participants would involve a random systematic sample of women who did not have career development while incarcerated and examine their rate of recidivism. The second subgroup would be women who did have career development while incarcerated and choose a random systematic sample to compare.

I don’t know how I would consider disproportional stratified sampling for this particular research question. Cluster sampling also doesn’t seem applicable to this target population as I believe there would be adequate records to pull from. If that were not the case, I think a cluster sample could be chosen from several in state prisons which have career development programs in place. The problem with this is that the results most probably wouldn’t be generalizable. The population chosen might have characteristics that are truly localized and not applicable across a larger population.

Convenience sampling in this case could involve access to participants at one particular facility; for example, I could perhaps be allowed access to the women in the career development class I facilitated here locally. It would be convenient because of location and because I know the people I would need to approach. It would also be very cost effective as it is close by and a small facility, so my sample size would be manageable. This would be a poor choice for this particular research question however. The results certainly wouldn’t be generalizable as there would be too many variables in other institutions that would make this one in particular not representative of a larger group.

Quota sampling would be a better choice than convenience sampling in terms of its ability to be generalizable. I might choose several similar facilities across the state of Virginia in different geographic areas, urban, rural and suburban for example. I would assume that the locality might be a variable that affects the outcome of the prisoner’s ability to utilize what they learned in class so this might provide a more generalizable target population to study while still having the advantage of relative low cost and convenience.

I don’t really see the use of purposeful sampling in this particular type of study. I’m sure that choosing successful former prisoners and those who have returned to prison multiple times would be indicative of the efficacy of career development programs. This method in this particular case would only seem to explain the phenomenon on a very individual level and not be at all generalizable.

I believe that proportional stratified sampling is the way to go here. I would need groups of participants randomly chosen from the following subgroups: women who are former prisoners who had career development in prison; women who had career development in prison, were released and then returned; women who were in prison, had no career development and were released; and, women who were in prison, had career development and did not return. I could access large numbers of data source from various low level prisons across a large geographic area to produce a large sample that would lead to generalizable results.

Steps for a literature review:

1. Select a topic    
   1. My general topic is career development workshops in correctional facilities
   2. Write research problem or hypothesis
      1. My qualitative problem is: “How does vocational skills training in prison affect the recidivism rate of women?’
      2. A. I searched Ebscohost for keywords correctional facilities and career development and sorted through the results for articles related to my topic.
         1. I searched ERIC thesaurus for different terms for career development and found career education/ career exploration/ career guidance/ and career planning. I added those to my search and chose related articles.
         2. Through my search I noted the names of several journals that might be useful to search later: Journal of Correctional Education; Journal of Career Education and Career Education Journal. Also took note of “Project IMPACT” which is a funded project related to my topic and CTE( Career and Technical Education- both worthy of further exploration.
         3. This search will be repeated using several other databases.
         4. Preliminary search netted 13 articles worth setting aside for further review.  From the abstracts I noted that some were primary sources and some were secondary. They will be sorted and the primary source articles will be read through. I will be looking first to see how closely they relate to my research problem and if they can be used in my literature review. Next I will  look for similarities and disparities in findings. I will try and determine what has been discovered about career development in correctional facilities and its affect on recidivism. I will also try and determine what is missing from current research and the direction my research might take me.
         5. I will read through the secondary source material for further refinement of terms that  and supporting ideas that I can search that might add to my knowledge base.
         6. I will organize the information in the relevant articles and write a review  incorporating a general overview of the articles (primary sources) and including similarities and disparities. I will note what information seems to be missing from current research and direct my research focus there so that I can add to the current knowledge of the topic.

I need more practice thinking like a blog. I still had to start this topic on paper and scratch through different ideas as I went. Something about scratching through seems to help my thinking process and I need to learn the blog equivalent. I will attempt to recreate the development of my ideas here.

The research area that interests me and with which I have some experience is career development programs in a correctional facility. I worked over two semesters with women in a local detention center facilitating a career development class. The piece that was always missing for me was: are the skills that we are hoping to foster in women who were soon to be released effective in helping them make better life choices and in helping them become prodeuctive, self-supporting individuals.

Ideas for stating  quantitative research study problem:

Is the rate of recidivism lowered in women who attend career development workshops in the correctional facility? I thought the participants should be better described so:

Is the rate of recidivism lowered in incarcerated women who attend career development workshops? I thought this wasn’t clear enough. When did they participate- before, during or after incarceration?

Is the recidivism rate of women who participate in career development workshops while incarcerated lower than the women who do not?

I believe this research problem contains the necessary criteria: it is researchable; it is of practical importance; it indicates a study of the cause and effect relationship between career development workshops (independent variable) on recidivism rate (dependent variable); it specifies the population to be studied; it is clear.

What makes this a quantitative research problem? It will show the relationship between career development workshops and recidivism and from that show the effectiveness or not of the objectives of these types of programs. This type of research will involve collecting lots of numerical data related to: career development programs available in correctional facilities; the number of participants; and the rate of recidivism in participants and non-participants. The collection of data will be structured, formal and specific most likely involving the analysis of inmate and released prisoner records from the prison and from parole officers. There will be a large number of study participants to be able to extract the most meaningful and accurate statistics. The researcher will remain detached and will be personally uninvolved with the particpants as it will be records and data that is studied, not the individuals themselves. The setting will be controlled as it will most likely be in the prison office.

The related qualitative research problem:

How do career development workshops in prison affect the recidivism rate of women prisoners? This may be too specific.

What is the influence of career development workshops in prison on the recidivism rate for women prisoners?This may be too specific, I’m not sure.

What is the influence of career development workshops in prison on women prisoners post release?

I believe this problem as stated contains the necessary criteria: it is not too specific or general; it will be relativiely easy to change the problem as information is collected; it does not imply any bias on the part of the reasearcher- influence can be positive or negative; it lends itself to description; and, it encompasses the question, participants and their setting- outside the prison.

What makes this a qualititative study? This is not too specific as it does not anticipate the result, nor is it too general; it allows space for describing multiple realities; it does not impose a certain structure and would allow for flexibility in design; it would entail using a limited number of particpants in order to collect long term data that involves active researcher participation with the particpants to understand the influence from the participants point of view; it would occur in a naturalisitc setting where the researchers could have the opportunity to observe the influence of the workshops by seeing the particpiants lives and hearing their stories.

I am a pinja!

I can see the value of both types of research in this particular case depending on the end purpose. Certainly a quantitative study would be helpful in providing a snapshot of whether the programs are worth pursuing especially when compared to other types of programs in that setting. It would be valuable in a bottom line type of discussion where future funding of these programs might be an issue.

The qualitiatvie study could be enlightening in terms of whether the program is worth pursuing as a whole, but also might provide insight as to what parts of the program are valuable to the participants and what parts aren’t helpful. The various components could be examined and compared to the objectives to see if the program needs changes to be more effective or meaningful to the participants.

This is the first class I’ve taken where group interaction has been a significant part of the learning process. Although every class in the School of Business involves a group project, the general format of the class is teacher focused. This has been difficult for me to adjust to. I still have this little voice in the back of my head that teacher knows best.

It has been interesting to examine that thought alongside the readings we have done about the adult learner and the wealth of knowledge and experience that each learner brings to the classroom. At the beginning of the semester I agreed with that statement about adult learners while at the same time thinking ” but still… ” The work we have done in class examining different learning styles and examining our own experiences with learning has helped me look at things a little differently. This has certainly been furthered by the work that we have done in groups to share ideas and the things we’ve learned about ourselves-what we’ve read about adult learners is true! The process of working in the groups and listening to others’ thoughts and ideas helped all of the readings become more than just ideas and words on a page for me.

I am hoping to have the opportunity to work again at the Detention Center next spring and be able to put into practice some of what I’ve learned. In some ways I was disappointed not to have the opportunity to do so while I was taking the class so that the ideas were “fresh”. Perhaps with more time to reflect on all that we’ve covered and how we’ve gone about it, I will be more prepared to use the information in a skillful way. Certainly there was a lot of theory covered that will be useful as I look at the detainees in a different light. Learning styles and ‘ways of knowing’ will figure prominently in my approach to teaching there. That information alone has made me look at my experience with incarcerated women with a much deeper understanding of their way of seeing things. I am hopeful that this new understanding will allow me to find better ways to adapt the material for the classroom, to reach them more effectively and to support their learning in a way that is meaningful to them individually. It will be interesting to try and incorporate a more group focused, less teacher focused class environment. We always encouraged the students to share their stories and listen to each with the hope that they could learn from each other. Sometimes it was successful, but the basic dynamic was certainly teacher as authority. Helping them to find their own voice will be an important step for some of them.

Although I don’t have explicit teaching responsibilities in my work life right now, our classwork on learning styles has made me step back now and then and think about how each of us on our work team is looking at a task or problem and the different ways we approach finding a solution. It has been fun to see our differences in action and interesting to watch the process of finding solutions together in a way I didn’t see before.

This has been a very difficult semester for me in that my attention has been truly divided and I have had demands that have kept me from participating to the degree I normally would. I will take some time over break to go over what we have done and use it to plan how best to accommodate a new schedule in the spring in terms of what I need as a learner and how to best contribute my strengths to the classroom. This class has certainly provided much food for thought.