Haven't found the Essay You Want?
GET YOUR CUSTOM ESSAY SAMPLE
For Only $12.90/page

The Power and Influence of Advertising Essay

1. Introduction

The questionnaire has been designed to record respondents’ views on the power and influence of advertising in various media.

The questionnaire consists of six questions, designed to collect both quantitative and qualitative data. Five out of the six questions (questions 1, 2, 3, 5 and 6) are closed questions, suitable for the recording of quantitative data. Respondents are offered a selection of pre-written answers and choose the category or categories most suited to their profile or opinion.

Questions 1 and 2 record respondents’ basic personal details, whilst ensuring that each respondent remains anonymous. Question 1 records the respondent’s sex, and produces nominal data. Question 2 asks respondents to select an age category. This is a preferable practice to asking respondents’ exact age, as people are occasionally unwilling to reveal this information, thus resulting in missing data. This question records ordinal quantitative data.

Question 3 asks respondents to choose which form of media they consider to be the most powerful. The questionnaire allows a choice of three responses; “still image (magazines and newspapers)”, “still image (billboards)”, or “moving image (TV)”. This question records nominal quantitative data.

Question 5 asks respondents to record whether or not a skilful advertising campaign would influence them to vote for a particular political party, purchase a particular product or brand, or revise their opinion on a specific issue or cause. Respondents are able to select as many or as few options as they wish. This question also records nominal data, as no chosen option is of a higher value than the others.

Question 6 records ordinal quantitative data by asking respondents to gauge their opinion of how much they feel that they are influenced by advertising, in comparison to other people. Respondents have the option to choose from three options; “less than other people”, “more than other people”, or “about the same as other people”.

In contrast, Question 4 is an open question, designed to allow respondents to elaborate on their perception of the power and influence of the specific media focussed on by the researchers. Naturally, this type of in-depth response contains far more detail, allowing an insight into the individual respondent’s feelings and opinions on the issue. The data provided by this type of question is also more difficult and time-consuming to analyse, as the analyst must sift through each response in detail.

Qualitative data of this type could possibly be allocated to separate categories to allow a quantitative analysis, but this would lose any subtlety inherent in the detail, and the categories would offer a fairly crude interpretation of the responses.

2. Design Issues

Although each researcher has asked all respondents exactly the same questions, there are minor differences in the design layout of each researcher’s questionnaire.

Researcher A lists all available options for each question on the questionnaire paper. He then records all quantitative responses with a YES or NO beside each respondent’s choices.

Similarly, Researcher B also lists all available options for each question on the questionnaire paper. She then circles all respondents’ selected quantitative responses.

Researcher C adopts a slightly different approach to recording his data. Except in the case of question 5, a multiple answer question, where all potential answer options are displayed on the questionnaire paper, Researcher C simply records each respondent’s answer to a single choice question below the appropriate question. Researcher C also attaches an alphabetical annotation to the available choices in question 5, apart from in the case of questionnaires 14 and 15.

A survey designed to obtain data on the power and influence of advertising in various media could possibly be made more inclusive by including a question on advertising via the world wide web.

3. Results

Each of the three researchers obtained 15 completed questionnaires.

All quantitative and qualitative responses for each researcher’s questionnaires are detailed in appendices 1 (quantitative data) and 2 (qualitative data). A summary of responses, attributed to each researcher, is given below.

3.1 Quantitative Data

3.1.1 Sex

Each researcher has gathered data from both male and female respondents.

Researcher A’s responses are heavily biased towards female respondents; 3 males (20%) as compared to 12 females (80%).

Both researcher B and Researcher C have achieved more of a balanced sample, with 6 male and 9 female respondents each (40% and 60% respectively).

3.1.2 Age Group

The age range of respondents differs dramatically, depending on the researcher.

Researcher A’s respondents are fairly evenly spread across the age range, with almost half the sample (46.6%) above the age of 40, and just over half the sample (53.3%) below the age of 40.

Researcher B’s data also includes respondents from each age group, although the sample is heavily weighted towards respondents in the 18-25 group (53.3%).

Researcher C’s data is far less representative of the complete age range, only containing respondents from the 18-25 and 26-30 age groups, and here it is heavily biased towards the 18-25 group (73.3%).

This disparity of sex and age representation may have an effect on the subsequent responses for each researcher.

3.1.3 Most Powerful Form of Advertising

The moving image (television) is regarded by the vast majority of all respondents as the most powerful form of advertising, regardless of age and gender.

73.3% of Researcher A’s respondents cited this as their preference, compared to 66.7% of Researcher B’s, and 80% of Researcher C’s respondents. Perhaps significantly, the very highest percentage preference recorded here is from Researcher C’s respondents, who are drawn exclusively from the youngest two age groups.

Images in magazines or newspapers are cited by 2 of Researcher A’s respondents, and by 3 each of Researcher B’s and Researcher C’s respondents.

Gender and age do not appear to have any strong influence over the data relating to magazines and newspapers, although no respondents are aged over 40.

Researcher A’s respondents are both female, one aged between18-25 and the other aged between 31- 40.

Researcher B’s respondents include a female aged between 26 -30, and two males from the 18 -25 age group.

Researcher C’s respondents are all female, and within the 18-25 age group (unsurprisingly, as all of Researcher C’s respondents are drawn from the youngest two age groups).

Images on billboards are only mentioned by 2 each of Researcher A’s and B’s respondents; the two sets of data that contain a wider representation of age ranges. The power of billboard advertising may be affected by a respondent’s age, as no respondent who cited this option is aged less than 26.

Researcher A’s respondents are a female aged over 50, and a male from the 26-30 age group.

Researcher B’s respondents are a male aged over 50, and a female aged between 31- 40.

3.1.4 The Persuasive Power of a Skilful Advertising Campaign

Respondents are asked to state whether or not a skilful advertising campaign would have an influence on them in relation to three factors; voting for a specific political party, buying a particular brand or product, or revising an opinion in relation to a cause or issue. Researcher A’s and B’s sample groups identified voting for a specific political party as the factor most likely to be influenced by advertising (41.18% and 39.13% respectively). These are the two sample groups with the widest representation of age groups. 34.38% of Researcher C’s respondents identified this factor.

An equal percentage of Researcher C’s respondents (34.38%) also identified revising an opinion in relation to an issue or campaign, compared to 27.94% of Researcher A’s respondents, and 33.33% of Researcher B’s respondents.

30.88% of Researcher A’s respondents identified influence over buying a brand or product, compared to 27.54% of Researcher B’s, and 31.25% of Researcher C’s (the group containing the youngest respondents) samples.

3.1.5 The Level of Influence as Compared to Other People

The question of how much a respondent feels that they are influenced by advertising in relation to other people varies quite considerably by each researcher’s data.

80% of Researcher A’s respondents felt that they are influenced about the same as other people, whereas none felt that they were influenced more.

Over half (53.3%) of Researcher B’s respondents felt that they are influenced about the same as other people, as opposed to 26.7% who felt that they are influenced less, and 20.0% who felt that they are influenced more than other people.

Researcher C’s respondents are evenly split (33.3%) across all of the three options.

Having said this, it is not possible to draw any meaningful conclusions from the data, as the sample sizes are extremely small and unrepresentative. These are, probably, all convenience samples (the researchers have delivered the questionnaire to colleagues, friends and family who they meet during the course of their weekly routine). It would be difficult to make any comparison between the three separate sets of survey data, as each researcher has obtained data from slightly differently weighted samples, as regards gender and age range.

3.2 Qualitative Data

In the case of the qualitative responses, there is again variation between each researcher’s data.

Researcher A’s qualitative data is extremely brief, mainly consisting of fairly short one sentence responses.

The qualitative data obtained by Researcher C contains several slightly longer responses but, again, the majority are rather brief.

The majority of Researcher B’s qualitative data responses are far more detailed. She has allowed her respondents to elaborate in their answers, recording the detail, resulting in potentially richer data for analysis.

3.2.1 Moving Image (TV)

As discussed in section 3.2 above, the moving image is cited by an overwhelming majority of all respondents, from all three surveys, as the most powerful form of advertising.

Researcher A’s respondents gave various reasons for this:

• Watching TV unconsciously enters your brain when relaxing.

• More real than still pictures.

• Music and talk-over help advertising.

Researcher B’s respondents elaborated in more depth:

• I think it is because you start to desire that object, because you can see it in full detail, and you can see how it looks in 3-D. I think it is the most powerful way of advertising because generally people watch television every day, so will get to see these adverts more so than having to buy a magazine or drive by a billboard.

• My eyesight is not as good as it used to be, so I like to be able to hear the advert and what is being advertised. I like to sit down and watch television after I’ve been at work; it helps me to relax and I think that helps me to be drawn into the advert.

Researcher B’s respondents show an awareness of the strategic placing of advertising within the television schedule:

• Because someone is talking to you, you are more likely to pay attention to the television because you are waiting for your programme to come on.

• Because certain companies advertise their products at different times according to how popular the programme is on television.

Researcher C’s respondents highlight the passive role of the viewer:

• You get the advert’s message where all you have to do is watch.

• TV is entertaining to watch and so you’re more susceptible to the ideas adverts present but you don’t necessarily notice it because you’re being entertained.

• Because television is more easy to pay attention to than something like reading.

Researcher C’s respondents also pinpoint the effects of advertising via an animated medium:

• It is interesting to watch television because theirs [sic] moving images and sounds that grab your attention.

• Because it uses sound and moving images to create an experience.

3.2.2 Still Image (Magazine or Newspaper)

Still images in magazines and newspapers are the second most popular choice for respondents. Researcher A’s recorded responses are extremely brief, and therefore not particularly enlightening:

• Persuasive form of advertising.

• Colours.

Researcher B’s responses are far more informative: Respondents state that:

• I think still images capture a moment or a feeling, and if you relate to that it can be powerful, whereas an advert can get lost in moving images, as there are so many more frames, plus a still image can really grab your attention.

• I think magazine adverts are quite powerful because you get to look at the picture and understand what is being advertised and what the purpose is for the advert. Also, in between lectures I read a lot of magazines so I get to see loads of adverts. Magazine adverts are nicely laid out and the colour helps sell the item.

• I tend to read a lot of magazines; it’s more to do with the fact that the images are stronger. I tend to look at pictures more than read articles. They have started to use black and white images again to make the product more powerful, and it catches your attention.

Researcher C’s respondents add:

• You are forced to look at the image because you’re already reading the magazine and, unlike TV, you can look [at] the images for as long as you want to.

• The way they use images to provoke certain thoughts. Often they use sex to sell products.

• The use of colourful imagery means that you pay attention.

3.2.3 Still Image (Billboard)

Unsurprisingly, billboards are selected not only for their imposing physical presence, but also by several respondents who mention travel as part of their daily routine:

• I’m often on public transport… they’re almost subliminal. I look at them because I don’t feel it’s as in my face as, say, TV.

• The adverts on billboards are very big and easy to read. I do a lot of travelling, so I am on the go all the time, so I pass many each day. When I am stuck in traffic it gives me something to look at.

• Billboard adverts are very powerful because they are on large scales, which means that it stands out… billboard adverts are big, colourful and bold.

A fourth respondent chooses this option, but states that: I don’t really like them… Appendix 1: Quantitative Data

1. Researcher A

Of the 15 completed questionnaires, all respondents answered all questions. Therefore, there is no missing data.

Questions:

1.1 Are You…

• Male3 (20%)
• Female 12 (80%)
[pic]

1.2 Which Age Range Do You Fall Into?

• 18-25 2 (13.3%)
• 26-30 3 (20%)
• 31-40 3 (20%)
• 41-50 5 (33.3%)
• 50+ 2 (13.3%)
[pic]

1.3 What Kind of Advertising Do You Find to Be the Most Powerful?

• Still Image (Newspaper or Magazine) 2 (13.3%)
• Still Image (Billboard) 2 (13.3%) • Moving Image (TV) 11 (73.3%) [pic]

1.4 Are You More Likely to Do Any of the Following as a Result of a Skilful Advertising Campaign?

• Vote for a Particular Political Party (41.18%)
• Buy a Specific Brand or Product (30.88%)
• Issue or Campaign (27.94%)
[pic]

1.5 Do You Think That Advertising Influences You …

• Less than Other People 3 (20.0%)
• More than Other People 0 (0%)
• About the Same as Other People 12 (80%)
[pic]

2. Researcher B

Of the 15 completed questionnaires, all respondents answered all questions. Therefore, there is no missing data.

2.1 Are You…

• Male6 (40%)
• Female 9 (60%)
[pic]
2.2 Which Age Range Do You Fall Into?

• 18-25 8 (53.3%)
• 26-30 1 (6.7%)
• 31-40 2 (13.3%)
• 41-50 1 (6.7%)
• 50+ 3 (20.0%)
[pic]

2.3 What Kind of Advertising Do You Find to Be the Most Powerful?

• Still Image (Newspaper or Magazine) 3 (20.0%)
• Still Image (Billboard) 2 (13.3%) • Moving Image (TV) 10 (66.7%) [pic]

2.4 Are You More Likely to Do Any of the Following as a Result of a Skilful Advertising Campaign?

• Vote for a Particular Political Party (39.13%)
• Buy a Specific Brand or Product (27.54)
• Issue or Campaign (33.33)
[pic]

2.5 Do You Think That Advertising Influences You …

• Less than Other People 4 (26.7%) • More than Other People 3 (20.0%) • About the Same as Other People 8 (53.3%)
[pic]

3. Researcher C

Of the 15 completed questionnaires, all respondents answered all questions. Therefore, there is no missing data.

3.1 Are You…

• Male6 (40%)
• Female 9 (60%)
[pic]

3.2 Which Age Range Do You Fall Into?

• 18-25 11 (73.3%)
• 26-30 4 (26.7%)
• 31-40 0 (0%)
• 41-50 0 (0%)
• 50+ 0 (0%)
[pic]

3.3 What Kind of Advertising Do You Find to Be the Most Powerful?

• Still Image (Newspaper or Magazine) 3 (20%)
• Still Image (Billboard) 0 (0%) •
Moving Image (TV) 12 (80%) [pic]

3.4 Are You More Likely to Do Any of the Following as a Result of a Skilful Advertising Campaign?

• Vote for a Particular Political Party (34.38%)
• Buy a Specific Brand or Product (31.25%)
• Issue or Campaign (34.38%)
[pic]

3.5 Do You Think That Advertising Influences You …

• Less than Other People 5 (33.3%) • More than Other People 5 (33.3%) • About the Same as Other People 5 (33.3%)
[pic]

Appendix 2: Qualitative Data

Respondents are asked in question 3 to choose which forms of advertising she or he personally finds to be the most powerful, followed by question 4 (an open question) which asks: “What makes such images powerful”?

1. Researcher A

Question:” What makes such images powerful”?

1. I don’t really like them… (2) F, 5
2. I’m often on public transport… they’re almost subliminal. I look at them because I don’t feel it’s as in my face as, say, TV. (2) M, 2 3. Link with reality, similar with reality. (3) F, 4

4. Persuasive form of advertising. (1) F, 1
5. Watching TV, unconsciously enters your brain when relaxing. (3) F, 3 6. More real than still pictures. (3) M, 3
7. Colours and strong personality. (3) F, 4
8. Movement. (3) M, 5
9. Message. Visual stimulation. (3) F, 4
10. Proof of a product to some extent. (3) F, 2
11. Colours. (1) F, 3
12. Stays in memory longer. (3) F, 4
13. Speech and actions (3) F, 2
14. Music and talk-over help advertising. (3) F, 3
15. Camera angles. (3) F, 1

2. Researcher B

Question:” What makes such images powerful”?

1. I think it is because you start to desire that object, because you can see it in full detail, and you can see how it looks in 3-D. I think it is the most powerful way of advertising because generally people watch television every day, so will get to see these adverts more so than having to buy a magazine or drive by a billboard. (3) F, 1 2. Sometimes it’s the tune they play with the advert or the characters advertising the product. It tends to hold your attention more than the others. (3) F, 5 3. Because it gets your complete attention. You are sitting down and watching it, it’s not like you’re driving by it and can’t see the full detail of the advert. Also. Because having seen the product on a moving screen you then remember it more. (3) M, 5 4. I think still images capture a moment or a feeling, and if you relate to that it can be powerful, whereas an advert can get lost in moving images, as there are so many more frames, plus a still image can really grab your attention.

(1) M, 1 5. Because you get to see a lot of adverts in a short space of time. (3) M, 1 6. I think because you get the sound with the advert it helps to catch your attention more, whereas a still image, you don’t. (3) F, 1 7. Because certain companies advertise their products at different times according to how popular the programme is on television. (3) M, 3 8. Because you get to incorporate sound as well, so it is not just a frozen image. (3) F, 1 9. I think magazine adverts are quite powerful because you get to look at the picture and understand what is being advertised and what the purpose is for the advert. Also, in between lectures I read a lot of magazines so I get to see loads of adverts. Magazine adverts are nicely laid out and the colour helps sell the item. (1) F, 2 10. Billboard adverts are very powerful because they are on large scales, which means that it stands out and therefore it grabs my attention more than, say, little adverts in a magazine.

Also I get bored of watching adverts on television, so they need to be striking and billboard adverts are big, colourful and bold. (2) F, 3 11. I tend to read a lot of magazines; it’s more to do with the fact that the images are stronger. I tend to look at pictures more than read articles. They have started to use black and white images again to make the product more powerful, and it catches your attention. (1) M, 1 12. My eyesight is not as good as it used to be, so I like to be able to hear the advert and what is being advertised. I like to sit down and watch television after I’ve been at work; it helps me to relax and I think that helps me to be drawn into the advert. (3) F, 4 13.

Because someone is talking to you, you are more likely to pay attention to the television because you are waiting for your programme to come on. (3) F, 1 14. They use lots of colours and are appealing to the eye. You are more likely to watch television and pay attention to the adverts, whereas if you see it on a billboard or magazine then you are just glancing at it and not actually paying much attention. (3) F, 1 15. The adverts on billboards are very big and easy to read. I do a lot of travelling, so I am on the go all the time, so I pass many each day. When I am stuck in traffic it gives me something to look at. (2) M, 5

3. Researcher C

Question:” What makes such images powerful”?

1. You are forced to look at the image because you’re already reading the magazine and, unlike TV, you can look [at] the images for as long as you want to. (1) M, 1 2. The way they use images to provoke certain thoughts. Often they use sex to sell products. (1) F, 1 3. Because television is more easy to pay attention to than something like reading. (3) F, 2 4. It is interesting to watch television because theirs [sic] moving images and sounds that grab your attention. (3) F, 1 5. You get the advert’s message where all you have to do is watch. (3) F, 1 6. A lot of adverts are made to be enjoyable and entertain. (3) M, 1 7. Their [sic] more interesting than something in a magazine. (3) F, 1 8. It’s fun to watch. They use ideas that will appeal to certain people. (3) M, 2 9. It is interesting to watch. (3) F, 2

10. The use of colourful imagery means that you pay attention. (1) F, 1 11. Thoughts are presented to you particularly when you are waiting for a programme to come on, and so your [sic] forced to pay attention to them whereas reading it requires effort. (3) M, 2 12. Because it uses sound and moving images to create an experience. (3) F, 1 13. It’s like watching real life. (3) F, 1

14. TV is entertaining to watch and so you’re more susceptible to the ideas adverts present but you don’t necessarily notice it because you’re being entertained. (3) M, 1 15. TV is fun to watch. (3) M, 1


Essay Topics:


Sorry, but copying text is forbidden on this website. If you need this or any other sample, we can send it to you via email. Please, specify your valid email address

We can't stand spam as much as you do No, thanks. I prefer suffering on my own