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

I feel that this is a bit of a tricky question that no one really has an answer to, that is why I have decided to try and find out the answer to this question in the final blog of year one.

Psychology is the scientific study of the human mind and its functions: the mental characteristics or attitude of a person or group.  Everyone is unique and different, when people experience things they interoperate them in different ways.  For example, some people would love to see a spider while others will hate seeing one, some people love high adrenaline sports while other people may become terrified by them.  So if everyone is unique in the way that they experience things how can psychologists say that they know how people are feeling.  Therefore how can everyone be put into the same category when it comes to reading one another’s moods.

Uttal (2007) argues that the mind and its mental process are inaccessible: what we can see and measure (behaviour) is only an indication of what is happening in the mind.  What is actually happening in the mind is unknown due to everyone’s emotions and feelings being slightly different to one another’s.  Psychologists can make measurement scales for people’s happiness and emotions but whether or not these results can be seen as accurate is a different story.

One way that I believe that psychologists could understand peoples’ feelings better would be via operationalization: defining fuzzy concepts, making the concept clearly distinguishable.  Using this would make it easier for participants to understand what they are being asked, therefore they can give a more reliable and precise answer.

Like me do you believe that psychologists can measure anything and everything, or do you believe that they can measure everything but their results may not be 100% as they are still to break through and make these results more reliable and solid.

Uttal, W. R. (2007). The immeasurable mind: the real science of psychology. Amherst, New York: Prometheus books.

Is the term “approaching significance” cheating?

Results are referred to as statistically significant when it is unlikely that they have occurred by chance making them significant.  The level of significance for a hypothesis is the level of probability that the research results were obtained by chance.  So if a test had a p-value of .05 or less there is a 5% probability that the results are not due to chance.  The term approaching significance means that the p-value is not .05 or less but close to this number: approaching significance.  The term approaching significance can be false and misleading, as the experimenter may believe that if they were to do their experiments on a wider population that the results would definitely be significant; this being the reason why they chose to say that their results are approaching significance.  However, this may not be the case, if they were to redo their experiment they could just as equally be the chance that their results lie even further away from the significance value meaning that their results are no longer statistically significant and are no way near approaching significance.

Shaver (1993) stated that the phrase ‘highly significant’ is silly, and so is ‘approaching significance’, and that it should either be a yes no answer, not ‘highly’ or ‘approaching’.  Some reflective researchers (Carver, 1978; Meehl, 1978; Schmidt, 1996; Shulman, 1970) have recommended that statistical significance tests should be completely abandoned as a method for evaluating statistical results due to the tests being so frequently misapplied.

I do not consider the term a way of cheating but a message to the researcher to carry on with their research until they can make their findings more conclusive and valid, otherwise their results could be dismissed and seen as inconclusive due to their results not being able to be classed as either significant or not significant but near significant.  Then again this is my opinion for our level of research, but when it is used in the real world I do believe that more rules and restrictions should be in place to prevent such terms, and that it can be considered as cheating; an easy way out for the researcher.

Carver, R. P. (1978). The case against statistical significance testing. Harvard Educational Review, 48, 378-399.

Meehl, P. (1978). Theoretical risks and tabularasterisks: Sir Karl, Sir Ronald, and the slow progress of soft psychology. Journal of Consulting and Clinical Psychology, 46, 806-834.

Schafer, W. D. (1993). Interpreting Statistical Significance and Nonsignificance. The Journal of Experimental Education, 61(4), 383 – 387. Retrieved from: http://www.jstor.org/discover/10.2307/20152389

Schmidt, F. (1996). Statistical significance testing and cumulative knowledge in psychology: Implications for the training of researchers. Psychological Methods, 1(2), 115-129.

Shulman, L. S. (1970). Reconstruction of educational research. Review of Educational Research, 40, 371-393.

Should psychology be written for the layman or should science be exclusively for scientists?

For this week’s blog we need to consider psychology as a science.  Psychology is the study of behaviour exhibited by all people and the study of cause and effect relationships, therefore the findings from such research are beneficial to everyone.  This is why the question should be considered.  On one side of the debate it can be argued that the professionals that have done all of the research and published reports should not  then have to write it in simple terms for “the layman” after all of the hard work that they have done.  They can say that it is the job of people who know what the research report says to teach other people what it means: teachers, lecturers, other researchers and so on.  On the other hand it can be said that psychology should be written for the layman, as it cannot be beneficial to everyone if it is unable to be understood by people who have not yet mastered the specialised language or jargon of psychology.

When reading a psychology paper the reader is bombarded with psychological terminology and jargon.  This can cause the reader to become confused and overwhelmed by all of the new terminology that they may not have faced before.  Here I have found a link to an abstract where Turney (1996) says that it is the responsibility of the scientist to explain scientific findings in ways that a variety of people can understand.   However, the fact that the definition for these words can be found in many places such as at the back of books, on a website, or in a psychology dictionary, enables the people who are new to the field of psychology to easily access such resources to help them out.

If the psychological papers were to be written in simpler terms, making it a whole lot easier for everyone to understand and read, the size of the psychological paper would increase drastically, and some of the papers are considerably long anyway.  Doing this would mean that other researchers, psychologists and scientists who wanted to read a paper and do further research into the findings that had been published would have to spend more time reading the paper due to its increase in size.  This would have a negative effect on the rate at which fellow psychologists could do their work due to the new restrictions of how they would be able to write, due to the need for background information for people who are not in the field of psychology.

At present when any psychological paper is published it has to be done in APA  (American Psychological Association, 2011) format.  The APA was designed to advance scholarship by setting sound and rigorous standards for scientific communication.  However, it is only easy to understand if the reader is in the same field, and they have learnt what all of the jargon and terminology means.  If people who do not have an understanding of psychology decide to read up on a published paper, they can use other resources that are out there to help them to understand the new field they have taken an interest in.  It is the same with all new things; you have to learn how to master them before they become easier.

Overall I believe that psychology should continue to be written the way it is at present, there are plenty of ways that people can find help to read psychology if and when needed.  There are also many blogs and magazines (also has blogs and other content on this link) out there that publish psychological reports in a format easy for people who are not experts in the psychology field to read and understand.

 American Psychological Association (2011). Publication Manual of the American Psychological Association (6^th Ed.). Washington, DC: American Psychological Association.

Dean, J. (2007, November 13). Why We do Dumb or Irrational Things: 10 Brilliant Social Psychology Studies. Retrieved from: http://www.spring.org.uk/2007/11/10-piercing-insights-into-human-nature.php

Buller, J. D. (2005, September 10). The glamour of evolutionary psychology. Retrieved from: http://www.newscientist.com/article/mg18725161.700-the-glamour-of-evolutionary-psychology.html

Psychology glossary (n.d.). retrieved from: http://www.alleydog.com/glossary/psychology-glossary.php

Turney, J. (1996). Public understanding of science. The Lancet. 347, 1087-1090. Doi: 10.1016/S0140-6736(96)90283-4

Does “statistically significant” mean there is definitely an effect?

This weeks’ blog is going to be about statistical significance, due to the fact that the other week whilst in PITS I was just agreeing with the mac lab assistance when she was telling me which parts of my line graph were significant.  I aim to find out what statistically significant means, and to find out if it means that there is definitely an effect.  Finally I will be giving reason for why I believe the outcome that I have chosen.  I hope you like reading this weeks blog and maybe learn something from it.

Results are referred to as statistically significant when it is unlikely that they have occurred by chance making them significant.  The level of significance for a hypothesis is the level of probability that the research results were obtained by chance.  So if a test had a p-value of .05 or less there is a 5% probability that the results are not due to chance.

We cannot conclude that our null hypothesis is false when our results are significant; significance testing allows us to say very little about the null hypothesis.  A significant test statistic is based on probabilistic reasoning, which severely limits what we can conclude. (Field 2005, pp. 53-54)

This means that “statistically significant” does not mean that there is definitely an effect on the results from your study, and the population that the study is based on.  An example of this is the hypothesis that all ravens are black, this hypothesis is proven to be true after you observe all of the ravens at the London tower.  However, for this hypothesis to become true (have an effect) on the population, every single raven would have to be found and looked at to prove the hypothesis to be either true or false.  As this example illustrates, a sample can provide incomplete or limited information about the population that is being tested, mainly due to the samples of the population being too small.

Both type l and type ll errors can also affect the results which determine the effect: type I error; this is an optimistic error: saying there is an effect when there isn’t, in other words a type I error is when a true null hypothesis is rejected incorrectly.  A type ll error  is a pessimistic error: saying that there is not an effect when there is, quite the opposite of a type I  error: the null hypothesis has not been rejected  when it should have been.

As you can probably tell, I do not believe that “statistically significant” means that there is definitely an effect.  Other than the reasons that I have mentioned, I believe that not only do you need a large population to determine a definite effect on a population but the same test could also be carried out several times to confirm that there has definitely been an effect.  Hope that this blog has helped you to learn something this week even if it is only a short blog.  Feel free to comment away.

Field, A. (2005). Discovering Statistics Using SPSS. London: Sage.

Howit, D., & Cramer, D. (2011). Introduction to Research Methods in Psychology. Essex: Pearson Education Limited.

Shuttleworth, M. (2008). Experimental Errors. Retrieved 16 Feb 2012 from experimental resources:  http://www.experimentresources.com/type-I-error.html

Walker, I. (2010). Research Methods and Statistics. Hampshire: Palgrave Macmillan.

The difference between a case study and single case design

I am absolutely delighted that blogging is back.  For my first blog this semester I have decided to do it on the difference between a case study and single case design.  It seems like an interesting topic to blog about, and to be honest I’m not too sure what the difference is myself.  So hopefully not only will this blog help you to learn something but I will lean something too.

A case study is an in depth study on a single case, the study can either be on an individual or a group, used to seek out the patterns and causes for behaviour.  Most case studies are done in the clinical environment.  Almost every aspect of the participants life and history must be analysed.  Two case study methods are: prospective and retrospective.  Prospective being the method where the individual or group of individuals are observed for an extended period of time while for instance the progression of a particular disease is being observed.  The retrospective method involves looking at historical information: start with an outcome and then work backwards to see if any information on the individuals’ life can determine risk factors that contributed towards the illness occurring.  An example of a non clinical case study is Michael Bamburg (2004) who conducted a project investigating identity development in 10-, 12- and 15 year-old-boys.

The strengths of case studies are that they use a broad method, considering many sources of data when drawing conclusions and inferences about individual participants.  The limitations are that the data collection methods can differ from case to case, and the conclusions drawn will only apply to the single individual.

Like a case study a single case design (SCD) can also be applied to an individual and a small group.  As we learnt in this weeks research skills and methods class, the baseline phase in such designs has the following purposes: to measure the behaviour of the participant over time and prior to the experiment, then concluding whether the behaviour is stable or variable.  The second purpose is to predict what type of trend the behaviour will indicate (upper or downward) under a no treatment condition.  When an independent variable (treatment) is being introduced and withdrawn a A-B-A-B single case design is used.  A: the initial base line showing the trend and variability of behaviour.  B: the introduction of the treatment, allowing the effects of the treatment to be displayed.  A: the withdrawal of the treatment.  B: the option of whether or not to return to the treatment.  The need to return to the treatment will only be if it will benefit the participant or if it is to provide additional evidence of the effects that the treatment has.  If more than one independent variable were to be used then an A-B-A-C single case design will be used.  The first A, and B are the exact same as those on the A-B-A-B design (baseline and treatment 1). After this A: the baseline and C: treatment 2.

The ABAB design is one of the oldest and most widely used singe-case designs developed in behaviour psychology.  It was initially used in laboratory studies with animals (Sidmand 1960); however, as the applied behaviour analysis movement got under way (Baer, Wolf, & Risley, 1968) it became a prototype for applied behaviour investigation conducted in the natural environment.

One of the largest advantages of the single case design is the potential of developing a novel hypothesis.  The disadvantages of a single case design being that the findings from the case study cannot be generalized to a wider population.

So the difference between the two studies is that in case studies nothing has to be changed, the participants or group of participants are observed and or have their life history looked over to see if an explanation for the illness can be found.  Whereas in a SCD an independent variable (treatment) must be included and removed for any information to be collected.  Overall both of the studies can be used to inform future studies and interventions, they can also both be seen as important and useful as one another.

After looking over both case studies and single case designs along with their strengths and weaknesses, I have come to the conclusion that I believe that the case studies are the better of the two.  I believe this as case studies can gather substantial amounts of information about both individuals and groups of individuals, although the data that has been collected for one individual may not be considered the same for another due to the collection methods sometimes differing from case to case.  If case studies are carried out for a set amount of time on groups of individuals or individuals using the same collection methods, I believe that the data that has been collected can be used for future individuals.  A well-known psychologist that used individual case studies to develop almost all of his theories was Sigmund Freud.

Kratochwill, T. R., Hitchcock, J., Horner, R. H., Levin, J. R., Odom, S. L., Rindskopf, D. M. & Shadish, W. R. (2010). Single-case designs technical documentation. Retrieved from What Works Clearinghouse website: http://ies.ed.gov/ncee/wwc/pdf/wwc_scd.pdf.

Reynolds, C. R., & Fletcher-Janzen, E. (2007). Encyclopedia of Special Education. Retrieved from http://books.google.co.uk/books

Shaffer, D. R., & Kipp, K. (2010). Developmental psychology: childhood & adolescence. USA: Wadsworth.

The medias interpretations of scientific findings

As you all know this is the last blog of the year, yey!! The topic that I have picked to do from the two that we were assigned is the topic of God.  I chose this topic as I thought that it would be a debate on whether or not he/she is real, but, this is obviously not the case as this argument will probably never be resolved.  However the newspaper article title does say ‘nuns prove God is not figment of the mind’, this is what I will be looking into for this week’s blog.  For those of you that have not read this article and did the other article I will begin by doing a brief overview and then get down to the argument.

A functional magnetic resonance image (fMRI) study was carried out with the aim of the study being to identify the neural correlates of a mystical experience.  Carmelite nuns participated in the experiment with their brain activity being measured while they were subjectively in a state of union with God. The results suggested that mystical experiences are mediated by several brain regions and systems.  Phenomenological data also indicates that the nuns actually experienced genuine mystical experiences during the mystical condition.

The Telegraph used the title ‘nuns prove God is not a figment of the imagination’, this can be seen by some as an accurate title, depending on their beliefs of God and their view of the findings from the experiment.  Let’s focus on my opinion, that being that The Telegraphs title for this article is not accurate in relation to the data that they have been provided with.  I believe this, as the aim of the study was to identify the neural correlates of a mystical experience, not to prove whether or not God is or is not a figment of the mind.  I also do not believe that whether or not God is a figment of the imagination was actually proved, as the study did not look into whether the mystical moments were real or imagined to be real meaning that the subject believed them to be real although they were not.  However, I do believe that the study managed to justify its findings concluding the different areas of the brain that are used when mystical moments happen.  Enough information was also provided for the findings to be used in similar studies.

The newspaper article concluded that the study neither confirms or disconfirm the existence of God: this is contradictory to the title of the article, also that the study allowed the nuns to create a sensed presence of God.  None of the information in the article seems to help people to establish whether or not God is a figment of the imagination, the article just seems to make you question your thoughts on whether or not there really is a God yet again.  In relation to the data and information that The Telegraph was given I do not believe that they had any grounds to use such a title as this was not the purpose of the study, and their findings and conclusion did not in any way suggest any questioning of God’s existence.

Overall I believe that the article was inaccurate in relation to the study, as the two do not appear to tie together as you would expect a justifiable article and report to tie together when they are based on the same subject.  What is your opinion on this, do you agree or disagree with what I have said?

Understanding the public’s perception of psychology, as influenced by the media: ‘it must be right – they say so’

These days the newspapers are renown for misinterpreting scientific data that they have been provided with to write an article on, totally misinterpreting it and turning people against things that can in fact be good for you, as well as making people believe that things that are bad for you are actually good for you.

Do you agree with my argument that the papers make a mockery of psychology and science as a whole?  To help you to decide I will look at an example of a case where the newspapers have changed the stories in some ways not sticking to the facts.

As most of you will remember in one of our SWaC classes we were provided with an original article about GABA (neurotransmitter gamma-amino butyric acid) and two newspaper article.  I am now going to look at the original article and compare it with one of the newspaper articles, hopefully by comparing the two I will be able to demonstrate how newspapers misinterpret the information that they are given and add their own little twists to the stories.

The original article: states that impulsivity is a multifaceted personality construct associated with numerous psychiatric disorders.  Recent research allowed them to characterize four facets of impulsivity: lack of premeditation, sensation seeking, urgency, and lack of perseverance.  Magnetic resonance spectroscopy (MRS) was used to measure the concentration of the GABA.  More GABA predicted lower urgency scores.   From these results the following conclusion was able to be made: there is definitely a relationship between GABA and a wide range of psychological disorders.

The Sun’s article: stated that it was a nasal spray and a tablet had been made as treatment to stop men from boozing too much or fighting.  They also stated that this treatment would regulate the levels of the chemical that can cause men to behave badly, having acts of impulsivity; cheating and gambling, and acting aggressively.

Conclusion: The Sun has totally misinterpreted the original article that they were provided with, as the original article provided no data about any treatment being made, let alone a nasal spray or a tablet.  They did not specify on any of the facets like the Sun did saying that men can become aggressive drink too much and fight if their GABA levels are not correct.  I agree that more research should be carried out, but the Sun’s interpretations of the findings are incorrect in places and they have added their own twist to the story with the nasal spray.

Overall the sad reality about the paper misinterpretations of scientific findings are that most people just read the papers and agree with what is said instead of actually do more research into the article that is read to see if the facts in the article are in fact true or not.  This is why it is sometimes up to the people whose findings have been misinterpreted or misrepresented to write another letter to the newspaper stating where they have gone wrong.  Sometimes the response letter is posted, but if it is it is normally a smaller article that people can miss when browsing through the paper.  So at the end of the day it is up to the readers whether they believe the papers or if they will actually research the facts to check both sides of the story.

Structuralism vs. Functionalism

For this week’s blog I will be looking into structuralism vs. functionalism.  To do this I will be explaining how and when these two different schools of thought came about, the difference between them, and why one was rejected and the other one emerged. After looking over the two schools briefly I will be able to compare them both and decide whether or not I think that the rejection of one school to make room for another school was for the best.

Unlike psychology, all of the natural sciences have a unified approach.  A unified approach being a common foundation on which all of the various fields within a discipline agree.   Biology’s unified approach is the common descent theory, physics is the unified field theory, and finally, chemistry is the molecular bonding theory.  In the late 1800’s psychology emerged as a separate discipline, with structuralism becoming the first school of thought for psychology in 1986.  The structuralism school tried to find a unified approach for psychology, using the schools beliefs, but failed to do so and to the day psychology still does not have a unified approach.  Structuralisms aim was to make understanding the process of the human mind easier, to do this they would break down the human mind using subjective measures.  The structure of the human mind was also explored by analysing consciousness into component parts.

Structuralism was later rejected along with its mind-body distinction, elements study, introspection observations, and subjective measures.  Functionalism explains the human mind and the purpose of it using the objective measures, unlike the structuralism that used subjective measures.  The school also wanted to work out how and why the mind operates in certain ways, to do this they explored the functions of the human mind.  Structuralism tried to look at the mind by breaking it down into parts to make it easier to understand, but by doing this I think that they failed to achieve what they were trying to do as they made things harder to understand by breaking them down into separate things.  Whereas, functionalism used objective measures, and studied mental operations rather than breaking things down into parts.

The functionalist schools methods have proven better and more reliable than those of the structuralism school.  Although without the appearance of the structuralism school the functionalist school may never have come about, as it is past experiences that we learn from, and this is how we become better at the things that we can do.  Therefore without both of the schools I do not believe that psychology could have progressed as well as it has done.  What is your opinion on the matter?

Below is a quick link to a website that does a quick overview of the origins of psychology:

http://psychology.about.com/od/historyofpsychology/a/psychistory.htm

Is an eye witness testimony reliable enough in court?

There are many different psychological theories that look at the reliability of the eyewitness; I had briefly looked over these by doing a spider diagram, on all of the different aspects that the police have to consider when interviewing the eyewitness, but I cannot seem to get the spider diagram to appear on the blog so I have had to list them instead: age; occupation; stress; discrimination; event factor: duration, violence; confidence of the eye witness; estimations; and mental state.  As I cannot go over all of the theories in one blog, the theory that I have decided to look at is memories.

Memory is the process used by which we encode, store and retrieve information.  It includes sensory memory, short-term memory and long-term memory.  Bartlett discovered that eyewitnesses tend to rewrite history in some ways by trying to fit memories into existing systems the more difficult it is, the more likely it is that we will distort our memories in order to make them fit.  This could be because there are three stages involved in the operation of memory;

1. Input; sensory data is translated into a memory trace.
2. Storage; this may be temporary or permanent.
3. Output; memories are useless unless they can be retrieved through recall, recognition, reconstruction, reproduction and/or confabulation.

When the input of the memory goes through these three stages, the memories can become mixed up with other memories, or lost altogether, making the witnesses memories of the events useless.  Meaning that the when the information is needed from the eyewitness in court there lack of memory skills can lead to the criminal not having enough information against them to be prosecuted.

Due to all of the factors that affect the eye witness’s memory of events the police had to think of a better way to interview them, and they chose the cognitive interview technique.  The cognitive interview technique is used instead of the standard interview technique as it is believed to result in a thirty per cent improvement of the eyewitness’s recall of the events that they have witnesses, with no increase in the number of incorrect responses (Geiselman et al 1986).

The key assumptions of the cognitive approach are:

-      Behaviour can largely be explained in terms of how the mind operates.

-      The mind works in a manner, which is similar to a computer: inputting storing and retrieving information.

-      Cognitive psychologists see psychology as a pure science.

The cognitive interview uses the following techniques to interview the eyewitnesses:

-      Mental reinstating the context of the event, i.e. sounds, smells and feelings experienced during the event.

-      Asking witnesses to recall the event in various orders or in reverse order.

-      Asking witnesses to report absolutely everything, regardless of the perceived importance of the information.

-      Recall the event from a variety of perspectives, e.g. imagine what the scene must have looked like from the point of view of several characters there at the time.

Each of these retrieval mnemonics allows the witness to re-view the event without the interference of leading questions but forces them to scrutinise their memory record.  The technique aims to maximise the number of potential retrieval routes and to benefit from overlaps, hopefully triggering otherwise unrecalled details of the event.  Whereas the standard interview technique is based on the leading questions which could lead the witness to give a false answer due to the way that the question was set out when asked.

After looking over just one of the factors that causes the eye witnesses memory of events to become mixed up with other memories, or lost altogether, it can be said that eye witness’s testimonies can be false, inaccurate and misleading.  Meaning that an innocent person can be convicted due to mistaken identity and being in the wrong place at the wrong time (thank god for forensics).  Even with the help of the cognitive interview, I do not believe that an eye witness’s testimony will be one hundred per cent accurate: leading to my conclusion that an eye witness testimony should only be used in court if it is backed up by other evidence: forensics; CCTV; other eye witness statements and so on.

Thank you for reading my blog, please comment on whether or not you believe that an eye witness testimony is reliable in court.

References

Bartlett (1932) cited in Applying psychology to crime.  Hodder and Stoughton: London

Geiselman et al (1986) cited in Applying psychology to crime.  Hodder and Stoughton: London

Why is reliability important?

Before I can go over the reasons why I believe that reliability is important, I think that it is important to go over what reliability actually means:

‘’Yielding the same or compatible results in different clinical experiments or statistical trials.  In normal language, we use the word reliable to mean that something is dependable and that it will give the same outcome every time.’’

If for example I was to say that a teacher was reliable I would mean that they are excellent at their job day after day.

Now that I have explained what reliability means I can look into the major types of reliability: internal reliability, and stability over time or different measurers.  Internal reliability indicates how consistently all of the items in a scale measure the concept in question.  There are three different ways to test for the internal reliability: split half reliability; odd-even reliability; Cronbach’s coefficient alpha.  The score is based on half of the items correlated with the score on the other half for the split half reliability.  For the odd to even reliability: the score based on even numbered items.  For the odd to even reliability: the score based on even numbered items correlated with the score based on the odd numbered items.  Alpha reliability: a measure of reliability based on the average of all the possible splits into two equal sets of items.  Stability over time or different measures is used to see how tests and measures perform at a time interval or across similar versions of a test. To do this there is the test-retest reliability, and alternate forms reliability.  Test-retest reliability: the extent to which the measure remains stable between two different points in time.  Alternate forms reliability: the extent to which two equivalent versions of the test correlated.

Now that I have defined reliability, and gone over the major types I can start to go over why I believe that it is important: reliability in science is essential: if the results were not reliable then they would be inconclusive and useless.  Firstly, whilst carrying out an experiment you must make sure that you are doing everything step by step and as carefully as possible: keeping the variables the same every time.  If this was not to be done then the results may not be valid meaning that they are not reliable, as they are not the same every time.  So, the best way to make sure that your results are reliable are to maximise the inherent repeatability or consistency in an experiment.  If you wanted to maintain the reliability internally, you would use as many repeat sample groups as possible, doing this will reduce the chance of an abnormal sample group skewing the results.  You will be able to tell if your results are unreliable as multiple sample groups that are reliable smooth out extremes that your first sample group produces generating more accurate sreads of results: maning that if the results continue to be widely different then the results are unreliable, meaning that you have to start over with your experiment again as something has gone wrong.

So do you believe that reliability is important or do you think that we should just go with the flow and not be bothered about reliability and be more bothered about the experiment at hand?

References

http://www.experiment-resources.com/definition-of-reliability.html#ixzz1agKLwY3U

Introduction to Research Methods in Psychology – Third edition: Dennis Howitt and Duncan Cramer.

References