Comments: S2: Final set = pure joy

COMMENT 1: http://statstastic.wordpress.com/2012/03/22/should-statistics-be-written-in-laymans-terms/#comment-73

COMMENT 2: http://psud6e.wordpress.com/2012/03/25/statistics-in-journalism/#comment-103

COMMENT 3: http://afshinpsychology.wordpress.com/2012/03/25/looking-for-mr-effect-should-you-add-more-people-to-the-equation-or-should-you-throw-a-load-out/#comment-70

COMMENT 4: https://whoblogsstatistics.wordpress.com/2012/03/23/have-blogs-been-a-useful-learning-technique/#comment-55

Thank you Alex! It’s been emotional! :’)

Have blogs been a useful learning technique?

So we all know the research module shook things up this year by removing the weekly tests and introducing the new blogging system. Therefore, it seems only fitting that for my final blog I should weigh up whether this has been a worthwhile change. I would suggest that although interesting, thought provoking and ‘easy marks’ I do not think they have been a useful aid to learning the foundations of research methods and statistics that will help me in the completion on my degree.

Firstly, I want to start by saying why I believe blogs were chosen to replace testing. To me it seems the blogs were implemented to engage students in thinking about the subject, whether that be ethics, evaluation, statistic related issues or simply how research has narrowed it’s population by its use of ‘jargon’. The blogs have most definitely captivated students, for me personally I have thought about, the area of research in particular, in a much more in depth way and I am constantly formulating opinions and ideas based on blogs I have read.

However, this new found skill is promptly turned off my blogs and comments are done. Blogs require a limited amount of time to create, this time is usually highly intense, however it only lasts a short period. Once that moment has passed I breathe a sigh of relief and return to my own little world. Therefore, my main point of contention is that I am not learning about research and statistics, I am learning how to write a blog in a way that will get me the required grade. Once the blog is finished, I “forget”.

Now, I use “forget” lightly, I’m sure deep down the skill I used is bumming around in my subconscious, waiting to be tapped into, but I am also in no doubt, that to bring that skill to the surface will require me to relearn. Therefore, regardless of whether I completed blogs this year or not, my third year is STILL going to require me to recapture these ablites. I believe the main reason for this is the lack of rehearsal, it has been found that rehearsal substantially aids memory retention and ultimately learning by moving information from the short-term memory into the long-term memory (Atkinson & Shiffrin, 1968). However, we were allowed to pick any topic, this created a huge diversity in the subjects being written about and therefore I rarely built on the same skills, for some blogs and comments I would be reading/writing about ethics, for others I’m trying to get my head around correlation and causation or even reliability vs validity. Furthermore, regardless of skills learnt it created a hugely diverse content, luckily we had testing in year 1, the majority of the items I read built on items I learnt previously by testing, so that made things easier. This variation substantially hindered my ability to learn and retain the information being thrown at me. Perhaps future years could have to blog about a specific items each week, the blog could be anywhere within the realm of that item, therefore using the same skills, but at the same time keeping things fresh for the commenters, using the same item would in all chances aid rehearsal as although blogs could be different they’re likely to be based around the same subject.

I mention the use of different skills, by this I am referring to whether we’re reading a paper and critically analysing it, or adding to a description of validity, perhaps even trying to learn about a new terminology such as the “file draw effect”. All these skills could have to be used as once and therefore by the time I’m done, all that has been achieved is my ability to tick the boxes for the marking guidelines and I would usually be hugely stressed with my head about to explode with new information.

So, the blogs replaced the weekly tests, I’m not a fan of blogging, I even resent the fact of being coerced into having another form of cyber-space commitment. However, what really does produce the most long lasting effects that will ultimately help me with my degree; blogs or tests? Well, according to the Testing effect , testing enhances learning. Although many people are not a fan of testing, and I get why….you actually have to learn the subject rather than having a week to read and rewrite things you find on the internet so blogs seem easier. However in the long run, surely it would be easier if we did actually learn the subject, i think the graph below shows nicely what I’m trying to say. In the first condition, I’ve learnt how to tick the guideline boxes and done well, however when retested later it becomes clear I haven’t learnt as much as a I would have done had I actually been tested, this pattern continues, and by the time summer has passed and I’m in year 3…what then? I’ve got a nice blogging history that has taught me diddely squat.

So in conclusion, has blogging been a useful learning technique and built the foundations for me to nail my third year? I would suggests no. I’m sure there are many people out there who have enjoyed blogs, but I am not one of them, they’re too time consuming for the amount of information learnt. I do however feel blogging could be made more conducive to learning should the topics be specific each week, aiding rehearsal.

No comments simply saying, “I actually enjoyed the blogs”, “some people aren’t good at tests”. Particularly the latter, it’s a stupid argument that can be applied to most things, for example, some people (like myself) are just not good at not eating, hence the chunky appearance, what of it, doesn’t stop people raving about obesity does it?!? no. Don’t do it!

Comments S2: Wk8.

COMMENT 1: http://itsdefinitelycaela.wordpress.com/2012/03/04/quit-your-jibber-jabber/#comment-54

COMMENT 2:http://psucee.wordpress.com/2012/03/11/is-it-ethically-ok-to-use-internet-sources-as-data-for-qualitative-studies/#comment-51

COMMENT 3: http://statstastic.wordpress.com/2012/03/07/is-it-possible-to-prove-a-research-hypothesis/#comment-58

COMMENT4: http://itsafreudianslip.wordpress.com/2012/02/19/stardate-205319-weekly-blog-should-psychology-be-written-for-the-layman-or-should-science-be-exclusively-for-scientistsscience-b/#comment-46

Is there anything that can’t be measured by psychologists?

Having now studied various modules in psychology it appears there are very few variables that can’t be operationalised in some way. Therefore as long as everything can be operationaised then logically there is nothing that psychologist cannot measure, EXCEPT the construct itself.

Psychology tries to develop theories or explanations of human behaviour through objective observations. Certain variables allow direct measurements to be taken such as weight. However, much of psychology is based on internal concepts that are not directly measurable such as fear. It is these ’non-visible’ concepts that much of psychology is interested in.

The unobservable

Psychologist can measure something that isn’t directly measurable by developing a meaurement procedure for measuring an external, observable behaviour and useing the results as a measurement of a hypothetical construct, for example an IQ test is a operational definition for the construct intelligence. It is from measurements such as these  we draw meaning with reference to the behaviour of individuals. However, no matter how frequently psychologists measure increase in heart rate it can never be proven that this is a true measure of fear, all psychologist are really measuring is the collection of items they have used to operationalise the underlying item of interest.

This brings in the idea of validity and reliability, for an operational definition to be useful in measuring the hypothetical construct it must be valid, ie. measures what it’s supposed to, and it must be reliable ie. consistently maintains it’s function in routine circumstances.  For example the construct of self-esteem has been operationalised by the Rosenberg Self-esteem scale. This operationalisation has been found by many studies to be reliable and valid (Silbert & Tippet; Kaplan & Porkony). Having said this the above example highlights the fact the psychologist can never be 100% certain that their operationalisation is measuring what they thought it was (low validity) as Goldsmith suggests that this measure lacks validity and fails to account for various indicators that suggest low-self esteem. This highlights the very nature of science, that nothing is ever proven, no operationalisation can ever be proven to measure the exact construct it is aimed to.

Ultimately psychology is using a form of paradigm, that suggests that the current operational definitions used are a way of measuring the actual topic of desire. Currently the realm or paradigm psychologist are working within suggests that this form of measurements is satisfactory, however like all paradigms they are subject to shifts. In a few years or a few hundred years psychologist may have found a way to measure every part of the human mind (some sort of mind reading device….unlikely).

If only!

So the root of what I’m suggesting is that as long as psychology follows this paradigm of operationalising concepts then it would seem there is little psychologist cannot measure. However, the point remains that although psychologists can measure various operational definitions which are thought to represent the desired target it can’t be proven, as is the nature of science. Therefore, it would appear there are many things that psychologists can’t measure, and that is the constructs themselves, unfortunately they are bound to measure the operational definitions alone.

Comments S2: WK 5:

Comment 1: http://itsdefinitelycaela.wordpress.com/2012/02/19/fmri-a-kind-of-scientifically-assisted-telepathy-2/#comment-38

Comment 2: http://chocolateraisons.wordpress.com/2012/02/19/is-it-ethically-ok-to-use-internet-sources-as-data-for-qualitative-studies/#comment-54

Comment 3: http://psucd2.wordpress.com/2012/02/19/placebo-effect/#comment-44

Comment 4: http://psud2d.wordpress.com/2012/02/19/is-compulsory-participation-leading-to-poor-science/#comment-49

What makes a research finding important?

Well this is a a tricky question because the judgement on what is important is subjective, a ‘one mans treasure is another mans trash’ situation. I imagine to the researchers, every bit of research is important. However, I am going to look at this question with a wider gaze and focus on importance with reference to society and aiding people. I would suggest that a research finding is important if the findings change our understanding or knowledge and if they directly or indirectly aid the wider population. I’m not suggesting a piece of research should cause a paradigm shift, but it should discover something new, or in some way be relevant to the general public.

I would like to state there is a difference between research that is important and research that is simply interesting to know. However, if the findings are important for society, then ultimately they are interesting. The importance of a finding is directly linked to how interesting the finding is. Regardless of how interesting the findings is when standing alone, if the findings are important there will be many people interested.
I would suggest the key to important research findings is adding something new to society. Whether that be adding a new direction or extending a current bit of research. Having said this research that was previously conducted using blue circles may have concluded a significant outcome. I would contend that altering the circles to blue squares and discovering the same significance is not important. Although the substitution for squares is new, nothing new has been discovered and the findings don’t appear to add to our knowledge. We now know circles and squares result in similar findings, but in the grand scheme of things, this seems trivial.

However, this is where the area gets tricky. What if the same bit of research was actually trying to discover the disparity of presenting information in different structural formations to individuals who have learning disabilities as a result of brain trauma. Regardless of whether both square and circular information results in similar learning ability, the information can aid how individuals with brain trauma learn. Therefore, it appears the importance of the research findings are heavily dependant on the research question, because the research question can ultimately decide whether the findings are going to be new / relevant to the population or just interesting to know.

I would suggest if the research question is important, in that it is trying to solve a problem or add to research in a way which has an effect on society, then regardless of whether the results are significant or not the findings are important. If the findings are insignificant the number of different answers to the question are reduced, if the findings are significant the findings are aiding the creation of an answer. Bringing this back to the beginning, if the research is simply interesting and has no implications then it’s not important. However, if the research is important, it is also likely to be interesting. I am aware that previously unimportant research could lead to further research that becomes important, but then it gets frustrating the my head will explode…..not dissimilar to what happened when I realised this…….

 

So i guess what I’m trying to say is, there is no definitive way to gauge whether something is important. However, if I was going to take a stab at it I would contend the importance of research findings is largely dependant on the research questions, and whether that question is trying to discover something new, develop and extend further research or in some way benefit society 🙂

Comments (semester 2, wk 3)

Have a lovely weekend Alex 🙂

Comment 1: https://whoblogsstatistics.wordpress.com/2012/02/05/is-it-ethical-to-use-animals-in-experiments/#comment-21

Comment 2: http://chocolateraisons.wordpress.com/2012/02/05/why-is-the-file-drawer-problem-a-problem/#comment-44

Comment 3: http://psuc3c.wordpress.com/2012/02/05/should-psychology-be-written-for-the-layman-or-should-science-be-exclusively-for-scientists/#comment-5

Comment 4: https://whoblogsstatistics.wordpress.com/2012/02/05/is-it-ethical-to-use-animals-in-experiments/#comment-25

Is it ethical to use animals in experiments?

It’s that time again folks, another semester of writing ridiculous blogs….hold onto your hats. I’ve decided to ease myself back into these by starting with a topic that already has multiple views which…naturally…makes my job of writing a blog easier. So I’ll explore the different avenues of this argument in todays blog. From everything I have done in psychology thus far and my own personal opinion I would suggest that using animals in experiments is ethical.

Firstly, it is important to state  that the use of animals in experimentation has helped discover, produce and implement vast amount of research which has utimately changed the way humans live, and saved peoples lives. This is therefore an argument that would fall under, does the ends justify the means? Experimenters are aware that the procedures they inflict on animals are potentially painful and cause huge amounts of distress to the animals. However, it is suggested that experimentation on animals is acceptable if:

• suffering is minimised  – experimenters are instructed to follow the 3 r’s; Reduction, Refinement, Replacement. Reduction to ensure techniques are improved and data shared amongst the scientific community to reduce the number of experiments taking place. Refinement to ensure medical care and living conditions are better and experimentation is invasive. Replacement to ensure that if means other than animals can be used, they are such as, individual cells computer models or fully informed human volunteers.
• human benefits are gained which could not be gained via alternative methods. 

However, the opposition to this argument is fierce. There is a large population of individuals who hold the contention that experimenting on animals is wrong because:

• it causes suffering to animals
• the benefits to human beings are not proven
• the benefits of experiments to humans could be found via alternate methods

Lets expand on these points. It causes suffering to animals, the idea that animals can actual experience suffering is actually a fairly recent concept investigated by Alridge et al, 1991. Alridge suggests animals are capable of experience suffering just like humans. Therefore experimenting on animals is no different to experimenting on a human with brain damage who is on a similar mental level to the animals in experiments (Peter Singer, 1991) . However, a critical review of the literature into the suffering of animals suggests that, apart from apes and possibly dolphins, the claim that animals experience suffering is unsubstantiated (Bermond, 1967). It has been suggested the 95% of the animals that are tested are rats or mice, there are legal requirements in place that set minimum standards for the humane treatment of animals (Animal Welfare Act), however this excludes rats and mice which could imply they are not considered to experience suffering, as suggested by Bermond, 1967. Therefore a major argument that contends the use of animals meets intense objection.

Secondly, the benefits to human beings are not proven. It is true, many drugs that are tested on animals fail to reach the human trial stage. Jane Goodall, 1999, suggested that “animals have not been as critical to the advancement of medicine as is typically claimed by proponents of animal experimentation”. This casts a shadow onto the used of animals, if animals are being subjected to experimentation with little gain, can we still consider it ethical? William Carey (2002) would suggest yes, experimenting on animals is not used to show that drugs are safe and effective in humans. Instead, they are used to help decide whether a particular drug should be tested on people. If there are 3 possible drugs to cure a deadly disease, two of these drugs unfortunately kill all the rats the 3rd causes no harm. This 3rd drug would be used on a small human sample in a small dose. The benefits to humans in this hypothetical case are obvious, as a direct consequence of having tested those drugs on animals, many lives could be saved. Had these 3 drugs all been tested many humans who receive drug A and B could have died. In this case the testing on animals is the lesser of two evils. Furthermore, animal research has aided the research of 70% of the Nobel prize winners for medicine. Perhaps there is an argument to suggest no direct link between animal testing and human benefit. However, antibiotics and vaccines which have been developed because of animal testing ultimately have had a knock-on effect in furthering medical research and helping to sustain lives, which in my opinion is a huge benefit.

Lesser of two evils?

Next, the results of animal testing could be produced in other ways. I feel I have touched upon this above, a drug needs to be found suitable to be tested on humans, the use of animals is not to find out if there are horrendous side-effects but to find out whether a drug is suited to be tested on humans. Regardless of everyones individual stance on whether the argument that the use of animals is ethical I would have to assume most people will agree that the use of animals is better -if only slightly – than using humans. I have stated above the use of the 3 r’s when considering experimenting on animals, these are in place to help prevent unnecessary testing on animals, this would suggest that if it is decided testing on animals is required it is because there is no current data on the research area, or replacement of animals with cells or computer generated models would be useless, and therefore it is not possible to produce the results in other ways.

A final point I would like to make is with reference the argument of certain harm to animals and the potential harm to humans. The animals will inevitably experience discomfort, side-effects or death, this is almost certain. However, the pain that humans may face is only potential harm. The harm done to humans failing to complete the experiment is unknown because it is impossible to say how likely the experiment is to succeed or what benefits the results could be. I would suggest that  this is a pointless argument when deciding whether it is ethical to perform an experiment, until the experiment is carried out nobody can know the importance of the study and the benefits it could produce, and therefore suggesting that humans may not benefit but animals will suffer is pointless because science would a continuous circle of ‘what ifs’.

In conclusion, I believe the use of animals is ethical, and that in this case the ends justify the means. It is sad that animals must suffer and potentialy die and of course there has to be a limit to the extent to which this is true, the definite discomfort of animals has to be taken into account, however, this is why the 3 r’s I have mentioned have been implemented. As conentious as the subject may be, I definitely feel that the potential to save even a few human lives via research is a result that could rarely have a restraint put upon it.

Homework for my TA – week 11/12

Have a lovely Christmas Joe 🙂

1)http://psucfc.wordpress.com/2011/11/25/is-the-qualitative-method-less-scientific-than-quantitative/#comments

2)http://cashy3.wordpress.com/2011/11/24/is-it-dishonest-to-remove-outliersor-transform-data/#comment-43

3)http://psych31.wordpress.com/2011/11/15/covert-observation-needed-or-ethically-unsound-you-decide/#comment-38

4) http://thought3.wordpress.com/2011/12/09/love-can-it-really-be-studied-as-we-do-core-subjects/#comment-46

Is it possible to prove a research hypothesis?

My instinctive answer to this question based on my knowledge in this area is that we cannot prove a hypothesis is generalizable across time and cultures, we can prove that in this research something happened and that it supported our hypothesis. However, there is a big difference between proving something and providing supporting evidence for it.

The example which instantly springs to mind is the one about swans. It was thought black swans didn’t exist, every sample ever collected showed the existence of only white swans. Therefore, it was commonly thought that swans only existed in white, does the fact that nobody had ever seen a black swan prove they do not exist?….No……because…..

Tadah!! Black swans were found in Southern Australia around 1790. Now previously, based on all the knowledge and samples conducted it was found the swans were a species which only produced white feathers, as far as the hypothesis that swans are only ever white, is concerned it was commonly accepted as proven. However, just because no one had ever seen a black Swan, and only encountered white Swans doesn’t mean other variation don’t exist.

Essentially, this was the main point I wanted make on the subject. Just because all research at that present moment points to one conclusion, doesn’t mean that in the future based on different research the same conclusion will be drawn. The only way to prove a hypothesis is to systematically disprove every other antithesis, which is basically impossible because there is an infinite number of alternatives.

With reference to the above example; although it is impossible to prove a hypothesis, it is markedly easier to disprove one. For example; my hypothesis is ‘all swans are white’, all it takes is the finding of one swan with a different feather colour and the original hypothesis is void because not all swan are white….according to the above picture there is at least one black swan in the world, so not all swans are white.

Within science the process of disproving a hypothesis is called falsification and it has been used to cause many paradigm shifts of the years. When a hypothesis stands over a period of time it can adapt and become a theory. However, when evidence arises which proves the original hypothesis wrong a paradigm shift occurs, or a shift in view/opinion. For example the world was considered flat until someone suggested the world was in fact a sphere = paradigm shift.

I guess then, that a research hypothesis can be proven for that time period, but it seems pointless. To prove something suggests to me, that the only possible answer has been found, and for another answer to be suggested and have supporting evidence means that original answer is no longer proven. Therefore I would conclude a research hypothesis can never be proven, it is just a statement about what you would expect to happen, and the evidence you collect is either in support or contradictory. The hypothesis can’t be proven because in time further enlightenment and knowledge can arise which alters previously held beliefs. In research there always remains the possibility that you don’t or didn’t have an adequate measure of finding the answer.