Internet as a Source of Data

In the world that we live and study in, it is undoubtedly true that the internet is the supreme source of information of all types. It can be used as a source of learning about experiences, theories, research and much more and one potential result of this is that it is theoretically very possible to use the Internet as a source of data – in particular, as a source of qualitative data.

The Internet seems to many to be the perfect place to relay experiences, opinions and thoughts on any issue and as such, it is easy to see how it may hold much appeal as a tool for any qualitative researcher. That said, is the internet a safe, reliable source of data or does it have the potential to cause more problems than it is worth?

Although internet communities may prove to be rich sources for qualitative data, there are some major concerns when data is collected in this way. The first of which that I will discuss is informed consent. For your data to be used in an experiment, it is always necessary that you give informed consent, something that is more likely than not, not possible when data is to be collected from the internet unless the data is personally taken by the experimenter. If the data is taken from a forum or chat room or some other public domain, it is very possible that this consent will not be present and in many cases, people may not be happy with their posts being used in this way as although by posting on the internet we are allowing the world to see, it does not mean that we are willing for this information to be studied etc. When the data is coming from somebody under 18, it is necessary to obtain the consent of their parent or caregiver. On the internet, it is often the case that details such as age, race, gender etc. are unknown which means that there can be a breach in the rules surrounding informed consent.

Another concern with the internet as a data source is that on the internet people often act in a way that they would not in an encounter of a different kind. This may work in favour for the experimenter, e.g. by using anonymity to remove any social desirability bias, or against them, e.g. by giving data that does not generalise to real life as well as it should.

The internet does have other strong benefits though. The most obvious of these is the ability to connect with many other people far more easily than ever before. This means that a researcher can quickly obtain data from a wide range of sources and if the proper methodology is taken, much of it can be very effective and useful. Using the internet, researchers can instantly connect with people who otherwise may have remained unknown meaning that rich data can be collected about rare events and studied in much higher detail than would otherwise have been possible.

The internet does have its dangers but is far too useful a tool to be dismissed completely with regard to qualitative research. It is true that certain lengthy precautions ought to be taken when gathering data from this source but I believe that if these are taken, and the research is conducted properly, the internet can be just as good a source for data as it is for information of any other kind.

Comments

http://smmitch.wordpress.com/2012/02/05/why-is-the-file-drawer-problem-a-problem/#comment-61

The Advantages and Disadvantages of Case Studies

Is it ethically ok to use internet sources as data for qualitative studies?

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

SONA System – The best way to collect data for dissertations?

By this point in time we are all well acquainted with the SONA Student Participation Panel. Each semester every one of us begrudges the hours that we must spend sitting in a room, usually staring at a computer screen switch between a few different images while pressing one of two buttons – waiting for that blessed text to come up saying “You have finished the experiment”. This is something that we quickly come to accept as a part of our course that can be responsible for some easy, additional marks but does this method of ensuring that projects acquire a respectable amount of data mean that the data obtained is less valuable than it may otherwise have been?

 

Ideally, participants that are to be used in psychological research ought to be randomly selected though this is obviously not an easy task, especially when you have 200+ people seeking data in a city the size of Bangor. As a result, we are forced into using SONA. The problem here is that most participants turn out to be 18-21 year old Psychology students who sit these studies on a regular basis. This means that not only are the results less generalisable than they ought to be, but there is an increased chance of people not acting as they would normally. This is especially true in the type of study mentioned earlier. By this point of my degree, I have sat at least 10 studies where all I had to do was watch a computer screen and press a particular button depending on the stimuli presented. By now I dread these and the only thing I am concerned with in these studies is when I’m going to get out. As a result of this my attention and enthusiasm towards the task are reduced to practically nothing and this may effect what my data show. Furthermore, I have heard stories from people who go into a study with no intention of giving correct answers, they will sit there and just answer randomly until the study is complete – I don’t need to go into the problems this causes for researchers with outliers and faulty data and such.

 

But, what are the alternatives? No matter how many fliers are posted around offering incentives to do research, it is unlikely that unless participation is compulsory enough data would be obtained for every student to do their project. Sampling through social networking sites may work for some but this is also a poor sampling source and is not feasible for the situation that 3rd year students face when so many people are searching for data. It is possible to hire professional recruiters for research but once again, this is hardly practical for this situation.

 

To conclude, I hate SONA and having to do these studies feels like the bane of my existence when the end of the semester roles around but it is without a doubt, a necessary evil. Without SONA, 3rd year projects would be all but impossible to organise and I know that by this time next year, my beef with SONA will certainly be a thing of the past – especially when it is me that gets to subject students to the horror that is the psychological experiment.

Comments

http://mdscurr.wordpress.com/2011/12/06/computers-a-qualitative-researchers-friend/#comment-42

http://psud2c.wordpress.com/2012/02/05/should-psychology-be-written-for-the-layman-or-should-science-be-exclusively-for-scientists/#comment-31

http://statsblog2011.wordpress.com/2012/02/06/ethics/#comment-34

http://psucb0.wordpress.com/2011/12/08/is-it-possible-to-prove-a-research-hypothesis/#comment-38

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

Psychology is a subject that, by its very nature, can be sure to hold interest in at least some way to everybody. Psychology covers such a wide range of topics that it is inevitable that research conducted by many psychologists will have some application in the real world and to real people, whether it is to do with stress management, social situations, child-rearing or a whole host of other subjects.

 

Much of what is studied in Psychology can be used in a practical way as well as a means of furthering our understanding of the mind and how it works. It is for this reason that it could be argued that psychology ought to be written to cater for the ordinary people of the world as well as those trained in understanding it, such as ourselves. If what is learned through studying psychology is not written in a way that laymen can understand, much of it may as well be going to waste.

 

That said, there are many areas of psychology that if we were to try and make it understandable for the masses, would lose its credit as scientific research. For example, look at neuropsychology. We were told at the beginning of one of our neuropsych modules to look at some papers in this area at the start of the course to see how much of it we could understand – for me, as well as most others, the abstracts for the studies we looked at meant absolutely nothing. By the end of the course, things were much more understandable which goes to show that some things simply are not understandable by laypeople. Common knowledge on some areas of psychology is so sparse or misinformed that it would not be possible for scientists to write what they need to and make it understandable.

 

In conclusion, there is no reason for psychology research to be reported in anything but scientific language. Those who wish to learn of the results and implications from studies can do so through a number of means that are designed to cater to those without specialist/expert knowledge on the subject. There are so many sources for information written specifically to educate laypeople that there is no reason whatsoever to “dumb down” psychology generally – is that not the purpose of education?

Comments

http://psychblogld.wordpress.com/2011/12/07/the-significance-of-significance/#comment-27

https://joestatsblog.wordpress.com/2011/11/25/the-qualitative-method-my-thoughts/#comment-27

http://klbpsych.wordpress.com/2011/10/28/what-makes-a-research-finding-important/#comment-39

http://notwilliam.wordpress.com/2011/10/28/ethical-reviews/#comment-80

BPS Ethical Guidelines – Maintaining a high standard or stalling psychology’s progress?

The BPS guidelines govern every experiment that a British psychologist wishes to conduct with the aim of maintaining a high-standard of psychological testing while ensuring that research and methodology remains ethical. This surely sounds like a good thing, right?

 

What casts doubt on this in my mind is the fact that a number of historic psychologists would never have obtained ethical consent if they were to request it today. A number of cases come to mind with regards to this potential censorship of science – one of the foremost being the study conducted by Philip G. Zimbardo in 1971. We all know about the Stanford Prison Experiment so I won’t bother describing it but I’m sure most would agree that if this idea was proposed today, it would likely have been rejected on ethical grounds. While this would have saved the participants of the suffering they endured, the implications of the Zimbardo experiment are huge and help us understand many modern events in a way that without it, we would have been unable to do e.g. the incidents at Abu Ghraib

 

Another controversial experiment that gave an important insight in psychology was that of Watson (1920) in his “Little Albert” experiment. This study used unethical means to gain an insight into learned behaviour and this case, how fears are the result of conditioned learning.

 

That said, was the knowledge gained through these experiments really worth the suffering of those who participated? In the case of many participants in the above studies, their life was never the same and it caused severe psychological harm (breach of the current ethical guidelines). Other participants from the Stanford Prison Experiment claimed that the experience was enlightening even if it was scary too and said that they were still pleased that they took part – and this wasn’t just those who played the role of guards either

 

With the guidelines in place, there is always a chance of missed opportunity with regards to findings of research that was prevented due to its being unethical but I feel that this is far out-weighed by the harm it prevents. The participants of some major studies in psychology were mistreated in some way which is of course bad, and whether you think the ends (the research implications) justified the means (the suffering) in these cases is up to you but what about those who suffered for pointless studies that wielded no results that could justify the means used to get them? For every Stanford Prison Experiment and Little Albert that used unethical methods to find something substantial, there will have been many more experiments that used unethical tests for no proper reason and this is prevented by strict enforcement of ethical conduct.

 

*http://en.wikipedia.org/wiki/Little_Albert

The Qualitative Method – My Thoughts

The Qualitative Method – My Thoughts

 

With the recent lectures regarding the qualitative method I have thought a lot about whether this method is one that I feel I would use in any research that I may or may not conduct. The qualitative method is one that is not so commonly used in many of psychology’s fields though its use has been on the rise since the 1980’s (Rennie, Watson & Monteiro, 2002). This type of research aims to analyse data that is not of numerical form such as speech or writing and is collected and analysed through various means. The purpose of qualitative data is usually to come up with a basic theory regarding a certain subject which gives the researcher a basis for a hypothesis that can be tested using the quantitative method. Most analysis is collected through interviews or observation of some form then later it will be transcribed in detail by the researcher. These transcriptions may are usually written in particular styles (e.g. line-by-line or Jeffersonian) and must include notation of pauses, laughs etc. as these types of things are often essential to our proper understanding of verbal communication and make it far easier to interpret the meaning behind the words and the impact the thoughts are having on the interviewees.* Once transcription is complete the researcher must try and find themes and meaning within the interview by coding it. There are various methods of coding too such as the Grounded Theory (where the hypothesis is generated as a result of the coding process – relies on line-by-line transcription) and discourse analysis (an approach that uses Jeffersonian transcription and searches for patterns/badly fitting cases to find meanings in the words).

 

The main advantage of the qualitative method is its keen attention to detail which allows emphasis on different things depending on the circumstances. This allows the theories to be generated as a result of what is found as it can be used to collect information on a range of related subjects at once as long as a general research question is in mind. Unexpected phrases or themes may pop up in data collection that can completely alter the understanding of the research question up until that point which can be important for the discovery of new ideas and hypotheses to be tested using the quantitative method. Unfortunately, there are distinct problems facing anyone who wishes to use the qualitative method. This method is incredibly time consuming with interviews often lasting over an hour which must be listened to several times before a complete transcription has been made. The work behind this method is not particularly easy (in my opinion, it is far more difficult than SPSS) with skill being required at all stages. For example, an unskilled interviewer may fail to pick up on important themes that may hold any level of significance. It also requires skill and dedication for proper transcription and coding of data. Qualitative research often has few participants due to the amount of work involved. This means that results often have very low generalisability – something not helped by how ambiguous data that is collected may be. What I mean by this is that an interview may wield very different answers while using the same questions depending on a vast array of variables e.g. interviewer, day, mood, past experiences.

 

I definitely feel more inclined to use a quantitative method in any research I would carry out. The main reason for this is the the qualitative method just doesn’t seem to be definite and objective enough for me – it feels unscientific and as though it uses at least some degree of guesswork. Although in statistical analysis we must create constructs by which to gain numerical values for our variables, it feels as though the answers we get through this are more correct and applicable to others – more scientific and empirical.

 

*More information on transcription and notation: http://www-staff.lboro.ac.uk/~ssjap/transcription/transcription.htm

My Comments

Do you need statistics to understand your data?

Do You Need Statistics To Understand Your Data?

http://rhiresearchskills.wordpress.com/2011/10/07/do-we-need-statistics-to-understand-data/

Do you need statistics to understand your data?

What makes a research finding important?

The average person is likely to have a different outlook on the importance of a study when compared with us students or those with a scientific background. For many, most research and findings are presented through the media – e.g. the newspaper – and little actual information is desired or presented. For this type of person, the main interest in an article comes from what the research involves and the possible implications of the results. This obviously holds a lot of importance within the scientific community too. I think everyone would agree that a study into treatment of cancer that finds a treatment to be incredibly effective under certain conditions is more important than one that finds that people who burn incense at home are more likely to be relaxed people. That said, I believe the criteria required for a scientist to consider research important may include more than just this.

For a scientist, the first step in deciding whether research holds a lot of importance will be some investigation into the methodology used to obtain its results. The main question is whether the research was conducted in an scientific manner. This is “a method of procedure that has characterized natural science since the 17th century, consisting in systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses.”* This means that in order for findings to be considered important, they must have been conducted by measuring and experimenting on variables in order to prove/disprove a hypothesis.

If this criteria is complete, then a scientist would probably look at how well the results can be applied to the whole of the target population. Generalisability is a crucial quality for findings as it shows that they can be applied to other people. This can be improved by increasing the sample size and ensuring that sampling is as random as possible to try and gain a representative sample. If the results are highly generalisable then any findings obtained will be able to be applied to a larger number of people which may mean that it benefits more people.

Research findings should also be shown to have high validity. What this means is that the experimental method should make it obvious that it is the manipulation of an independent variable that causes the change in the dependent variable. This can be identified through signs of bias, confounding variables or just chance.

The effect size of results can also impact the importance of findings. The reason for this is that two studies can both show a significant effect, though one may have a very weak effect size while the other is much stronger. If this is the case, then it would be obvious that the stronger effect size shows a much more significant effect and is therefore probably a lot more useful.

Of course, one last thing that will always impact the importance of findings is how applicable they are to everyday life. Something that impacts and improves the way we live our lives in some way will always be considered as more important than something with no everyday use in my eyes.

 

*Oxford English Dictionary