Example of Research

We used an example of research looking at The Demographics of Dubstep.

To do this he planned to look at:

• Websites

• Record labels

• Artists

• Blogs

• Forums

• Radio stations

• Magazines


FIRST he asked himself questions- what did he want to find out?

Then he used WIKIPEDIA to get a brief understanding and SERPs (search engine results pages) to make a note of communities and popular sites along the way.

This browse of th genre's online prescence will give a scale of the demographic and establish who the popular online artists and communities are.

He also took into account related searches to see what is related online. 

Using 'Double click ad planner' he got stats that showed him how popular the related websites found were. 

He used backlink (incoming link) analysis tools such as Majestic SEO  to track how many links there are to these websites around the web

He then hunted for directories of dubstep sites then vetted them for relevance and sites that related to the style, such as urban clothes and merchandise and looked at how far these reached. 

All of this research led to him being able to make a conclusion about the demographics.  

More forms of research...

Research should be OBJECTIVE rather than SUBJECTIVE

Information based, not opinion based.


Experiments (testing a theory by researching)
+ easy to control conditions and closely track progress and changes
- difficult and expensive to do on a large enough scale to get reliable information
-people are not usually in their natural environment and know they are being tracked which may affect result
> Could be used for tests on products to see if they work like cosmetics.

Participant observation (researcher put in research activity)
+ Gives a close and direct opinion or very direct information. First hand primary research can seem more trustworthy.
- could also make observation less useful and in little depth if they are distracted by participating
- subject to bias
> Can be used for things like game testing where a researcher plays it with others and tracks other people's opinions

Historical research 
+ good historical research will be very useful as a background in the understanding of a product or media product. It could also be inspiring for those wishing to improve the product and highlight problems.
- observing behaviour and social changes can put product developments into context
- There is not always enough information available
>  Advertisers may use the history of a product to take on a nostolgic point of view in marketing.

Rhetorical analysis (analysing the media language and how it persuades media)
> for example persuasive scripting in an advert could be analysed and show a taget audience in a certain light
+shows what the media thinks of the audience and how consumers are directed
- audience could behave and react in numerous ways so it may not be particularly useful research in that it couldn't be reapplied particularly and is unspecific and uncertain. Generalisation.

Comparative analysis (Comparing different media products.)
+ can highlight what might be missing from a product
- for looking at things like genre and audiences
e.g. by looking at a product then working out the taget audience

Competitor analysis (analysing and rating competitors products)  
+ gives a fuller view of the market, what you're up against, what you need to be better than and different from.
- rating and assesing in this kind of analysis can easily subjective and bias
eg. comparing two music websites before you go and make your own

Questionnaires and their problems. (+ Interviews)

Well made questionnaires can be useful for quantifying relevant information. They can be used for evidence of generalisations that can be made and just how valid those generalisations are. However they are a flawed research method if not done right.

-  For information to be useful and generalisable you have to have spoken to enough people. 

- Questions must be carefully written to allow no room for confusion in those that fill them out. You must be careful that questions cannot be interpreted in different ways. 

- Possible answers should mostly lead to quantifiable information, not a huge array of answers which will be difficult to get vast information from. So questions should be specific with limited answers. For example you could use tick boxes with options of answers.

- If you want a written answer that is not quantifiable (perhaps in order to use direct quotes and have more depth in your findings) then you must be sure to leave enough space for the kind of answer that you want. So if you want one word answers leave space for that and if you want a paragraph leave space for that. 

- You should not ask a question that shows some sort of bias that may sway a persons answer

e.g. 'Do you steal music? '

The word steal has negative connotations and people are less likely to answer honestly than if you say

'Do you download copyright music by file sharing? '

- Questionnaires MUST be specific so you need to make your aims clear before you begin.

- You need to consider who is filling out your questionnaire and if they are the best group to target. You must also know how you plan to target them and be sure that there are questions that tell you exactly who has filled out the questionnaire and specific things about them e.g. age, gender, occupation etc. 

vs. One to one interviews

Doing one to one interviews can take out some of the problems that questionnaires have, but brings it's own problems. 

They quite simply take up more time, would be more expensive to conduct, regulate and standardise interviews. 

You can get far more information and much more depth from answers than you can with questionnaires while still being able to get some generalisable information.

However people may be less likely to be honest and there is more of a chance of possible bias in the questions from the interviewer or bias from the interviewers reactions that would effect findings.

Focus Groups

Using a trail focus group we discussed which e media advertising is most effective.

It is important when using focus groups to take note of the demographics of those attending, for example we had a group of  17-18 year olds with 80% female.   It's important to know your demographic as this will mean the findings can be more valid when put into context.

A leader runs the group, asking questions and engaging in conversation with all the individuals involved encouraging group discussion. This is to create a relaxed atmosphere where people will feel comfortable in discussing their own opinions. This can make focus groups more useful as it will mean people will bemore likely to fully express an opinion rather than perhaps say what they think an interviewer is going to say.

The group can have negatives as people are easily influenced by a dominant characters opinion or may change their opinion when hearing someone elses.

Audience figures

Audience figures are extremely important in the media because if you don't have an audience you don't have a show. Viewers mean money and it makes each programme worth the money it takes to make it. This also means that channels can fund and profit from shows with advertising and advertisers will pay an amount that depends on the amount of people watching.

Why and how does the media use research?

 MARKET RESEARCH
Target audience, competition and all their info, how many people will buy it, distribution logistics, funding and potential advertisers.

Good research sells a magazine, not to the audience but to the advertisers using...

Audience profiling-
[from IPC. com (eg from NME magazine)]
Stats cover:
-DEMOGRAPHICS
Gender, age, social class ranking (eg. ABC1) & occupation of target audience. Possibly orienta
tion.
-PHSYCOGRAPHICS
Often in lots of detail, naming and creating a person.
Personality of readers.
eg. "NME readers are influential's when it comes to mobile phones." and the knowledge from NME makes them "the authority in music in their peer group."
Values, qualities, attitudes, interests, beliefs.
eg. "They tend to buy a new DVD every month (higher than the national average)"
- Then circulation and readership sizes.


- Media companies mostly hire outside media researchers to do the research eg. Nielson


SOCIAL MEDIA RESEARCH
(i.e. Facebook)
[from accessed 9/9/10 Social media research article www. nma . co . uk]

- Which? estimates social media research costs a tenth of postal questionnaires, but of course is only of benefit to companies whose target audiences are in touch with social media.

- Near-instant feedback for minimal cost.

- Coca'cola are an example of a company who are actively using and engaging in social media for research
“The beauty of social media is that it’s so instant and, particularly for youth, you can talk directly to people using their language and where they increasingly socialise and live their lives,” -head of planning Beth Corte Real.
- Dell also use social media to get a better idea of what their consumers want
“The great thing about social media research is it’s two-way and instant,” she says. “...we feel it’s the perfect medium to ask questions and open the floor to suggestions.”
- Can be particularly useful for SMALL COMPANIES WITH FEW RESEARCH RESOURCES as social media research is simple and very cheap.

The article says " While face-to-face research, which can reach a wider demographic, is unlikely to be replaced, the interactivity offered by social media looks set to continue attracting companies of all sizes." though it could be said that as everyone gradually moves online that social media will be able to cover the majority of demographics and will eventually become one of the most used methods of research.

To summarise it is:
- Very quick
- Economically viable (cheap!)
- Two way

but

- doesn't reach all demographics
- can't tell you everything you need to know

QUANTITATIVE RESEARCH Practice - Are fashion magazines fairly representing ethnic minorities?

Using the primary data of British womens fashion magazines we observed the balance of imagery of white ethnicy and UK ethnic minorites.

This research was conducted in a quantitative manner, looking through magazines and simply counting how many people of each ethnicity were found.

RESULT:
In a 2004 issue of Vogue we found that all bar two (out of over 100) of the models in both adverts and articles were white. There were images of two black celebrities and one black catwalk model. Advertisements included only one non-white model who was Asian but only in the background.
Overall this brought us to the conclusion that in this particular issue of Vogue white women were over represented in comparison to the population statistically. However this will relate to it being from 2004, as there has been progress. This will also relate to the target audience, for example a Muslim Asian would not approve of many of the seductive images in Vogue, or the clothing therefore is not likely to be represented as they are not targeted.
Issues include the tiny scale of the survey which can not fully show the statistics of all fashion magazines and their representation. Also the date of the publication will effect the results. There is also little detail here about the kind of representation given so it would perhaps be more useful to take note of what each ethnicity is doing.

Media research methods: overview

Primary research
- Straight from the source. Institutions finding out info for themselves.
eg. Market research; Surveys, questionnaires, focus groups* or one on one interviews**, observation.
-Empirical research- any kind of observable research like a persons behaviour, rather than thoughts.
-Direct contact with producers of a media product, like a site, is a seldom used form of research that could be very useful.
- Chatrooms and forums could be a useful tool for research.


*(target audience chatting about a product and reviewing it, also used for films, online stuff...)
**(so you can't be swayed and can be really grilled)

Secondary research
- books, online research, newspapers etc etc etc

- wikipedia is a lazy version of online research and very unreliable!
- The Guardian Media (on Mondays) is v.v.v.good and The Telegraph technology section.
- The media show from radio 4 as a podcast is useful.


Combining research methods leads to fuller end products.