Aim and Hypothesis
Aim: to carry out a tourism investigation on the island of Ibiza
Null hypothesis: there is no variation in the type of tourism and its impact in the two contrasting resorts of San Antonio and Ibiza Town
Our null hypothesis was chosen because:
- It is objective, allowing us to approach the investigation with little bias as there are no presumptions about expected outcomes.
- It is based upon the aim of the study and is specific to the chosen study area of Ibiza.
- It is testable as it focuses upon variations i.e. types of tourists and impacts of tourism within San Antonio and Ibiza Town.
- This null hypothesis can be rejected at the end of the study if differences are found to occur from statistical analysis. We can therefore prove or disprove our hypothesis.
Location and Characteristics
Ibiza is located 87km off the east coast of Spain, in the periphery of Europe. It makes up part of the Balearic Islands along with Mallorca, Minorca and Formentera.
Ibiza Town is situated in the south east of the island, 13km from San Antonio in the west. San Antonio is an urban area which was originally a fishing village but has expanded with modern tourism, becoming increasingly commercialised with modern hotels and amenities catering to the 18-30 demographic that is attracted to the area. Tourist facilities in the area include bars and nightclubs, hotels, restaurants, beaches and attractions such as boat trips. It has 3 major areas: the Sunset *****, Playa d'en Bossa and West End.
Ibiza Town is also an urban area but is much less commercialised than San Antonio. It has a harbour to the north east and is split into two main parts: the old town Dalt Vila, which is a World Heritage Site, and a more modern part, Eixample. Generally, Ibiza Town provides a more cultural experience.
The summer climate of Ibiza is typically in the mid-20s, reaching 30 degrees in August. October receives the most precipitation with 69mm of rain falling.
Friedmann's core periphery model:
Illustrates that there are more prosperous core regions within Europe and levels of wealth, economic activity and development decrease with distance, so places in the periphery become increasingly poor. As an area develops, one of two process is likely to occur:
- Economic activity in the core continues to grow as it attracts new industries and services (banking, insurance, government offices), thus levels of capital increase and the region will be able to afford schools, hospitals, better housing and better infrastructure/transport. These pull factors will encourage rural-urban migration, which is also driven by push factors in peripheral regions (lack of employment opportunites, low paid jobs, limited government investment).
- Industry and wealth begin to spread out more evenly. Several secondary regions may develop, resulting in the decline in dominance of the original core. Although, there will still be peripheral areas that were less affluent.
If this is linked to the enclave model (which illustrates how certain areas are designated specifically for tourism and have little multiplier effect in the local economy), this would lead me to expect most of the tourists in Ibzia to be from affluent core countries such as the UK and Germany. People from these countries travel to the periphery in order to get better value holidays.
Tour operators and airlines will have their HQs in the core, with their subsidiaries in the peripheral areas, allowing them to maximise profits owing to cheaper wages and lower resource costs.
The Butler model:
Looks at the way tourist resorts grow and develop over time. It has seven stages: exploration, involvement, development, consolidation, stagnation and rejuvenation or decline.
Our initital data showed that Spanish tourism had peaked in 2007, suggesting that stagnation may be occurring, possibly due to the recession in 2008-09 which led to a drop in tourism. Furthermore its overpopularity with young adults, often in single-sex party groups, discourages more affluent tourists from visiting.
This is further enhanced by research into 'A Taste of Spain', a government policy set up in the hope that the cultural background and natural beauty of the island will be rediscovered by more mature tourists. The aims of the policy are:
- Increase the number of off-season flights
- Improve tourism services
- Boost awareness of the island's natural environment
- Increase provision for pensioners
It is hoped that these aims will lead to the rejuvenation of the area and ensure that Ibiza is no longer a budget holiday destination.
Health and Safety
A thorough risk assessment was carried out prior to the visit which enabled us to categorise and manage the risks associated with our enquiry. We classified likelihood and severity, giving them a score from one to three with three being very likely/severe. The product of these two results then determined whether certain aspects were low, medium or high risk. 0-3 was low risk, 4-5 medium risk and 6-9 a high risk. For example, the likelihood of getting sunburn/sunstroke was classified as a two and the severity of this was given a two, the product of these scores if four so this was categorised as a medium risk.
Other risks included:
- Theft (keep valuables in a safe)
- Air/road travel (wear seat belts, use reputable firms, listen to safety instructions)
- Harrassment as group of young females (work in groups and in daylight hours, teacher supervision, have mobile phones)
- Illness (eat in reputable restaurants, carry European Health Cards, school's insurance policy)
Every member of the target population has an equal chance of being selected.
- Removes bias
- Use is easy once the random numbers table is constructed
- Can be time consuming to obtain the random numbers table and then gather your results
Samples are taken at fixed intervals.
- Shows a more equal representation of our study area
- Gradual variations along a transect can be observed
- Bias is introduced as not every member of the target population has an equal chance of being sample which can lead to over/under-representation
This method is used when the parent population is made up of sub-sets of known size. These sub-sets make up different proportions of the total, and therefore sampling should be stratified to ensure that results are proportional and representative of the whole.
- If the proportions of the sub-sets are known, it can generate results which are more representative of the whole population
- It is very flexible and applicable to many geographical enquiries
- Correlations and comparisons can be made between sub-sets
- The proportions of the sub-sets must be known and accurate if it is to work properly
- It can be hard to stratify questionnaire data collection, accurate up to date population data may not be available and it may be hard to identify peoople's age or social background effectively
Carried out using systematic sampling. Completed three times at each location at different time periods, sampling for five minutes each time.
In Ibiza Town it was completed at 1, 2 and 5pm. We split into two groups with one group sampling at sites 1, 2 and 3 and the other group sampling at sites 4 and 5. Within these groups we stood on opposite sides of the road to count pedestrians travelling in each direction. After completing the sampling at one site, we would move on to the next.
In San Antonio we completed the sampling at slightly different times of 12, 2 and 4pm. Again, this was done in groups for safety.
- Straightforward method with no special equipment required
- Were able to measure the flow in both directions to see if there's a pattern
- Method is predetermined, no decisions required at the site
- No bias
- Disputes over whether to count a pedestrian twice were not settled prior to the fieldwork
- May be daily variations so a mean should be taken over several days
- Temporal issues in that the timings were different at the two sites
Data Collection (continued)
Random qualitative sampling. A bipolar survey was carried out at each site in both San Antonio and Ibiza town. It is based upon your own opinion and is therefore subjective. We picked a number from the random number and used it to determine where we would stand while completing the survey. For example, if the number 614971 was picked, we could use it to walk 6 paces forward, 14 paces right etc. Sites were scored from -3 to +3 against certain categories such as quality and upkeep of buildings; pavement quality: width, excrement, litter; high street chain stores, and services for locals.
- Easy to complete
- Can show variations in the quality of the environment of the two sites
- Used random sampling to limit bias in which areas were surveyed
- Opinion based/subjective (could have improved this by making our standards more consistent by making the criteria more detailed)
- Didn't use the random sampling method properly, and in some sites we abandoned it altogether
Data Collection (continued)
Land use transect
At site 1, with our backs to Dalt Vila, we sampled the land use of the streets for all buildings on the left hand side. We used squared paper and drew one box to represent 2m across and one storey high. Clear coding was used for each land use so we could easily see clusters, this is particularly evident in Ibiza Town with retail units and gift shops. All land use was sampled both at ground level and at height to illustrate the true land use of the area, therefore in our analysis we’ll be able to compare the data, including all storeys and just at base level. Predetermined criteria were used for land use to make it less subjective.
- Predetermined criteria allowed us to decide quickly what land use was in each box, for example, G represented gift shop, R retail and PH pub
- As we went along it became easier and quicker to carry out
- Limited equipment required
- People did not follow the criteria in the same way
- Key was limited, needed expanding in San Antonio where there was no category for clubs
Data Collection (continued)
The questionnaire used random sampling and gave quantitative and qualitative data. A random number was selected and we decided how to use it. For example, if 758312 was the number, we may decide to use the first number and ask the 7th person to pass us to answer our questions. Although this was the initial method, it was soon discarded owing to a lack of public participation. We were forced to ask anyone who was willing to help. At each site, 10 questionnaires were taken in total, meaning the group asked one person each. There were open and closed questions to ascertain the reasons for being in the area.
- Qualitative data = rich in detail and quantitative data = able to use in statistical analysis
- Had to abandon choice method of sampling
- Language barriers caused difficulties - could lead to skewed data
- Male population was sampled more due to a willingness to participate, particularly in San Antonio. Similarly, we were more confident in approaching elderly people so our data may over-represent this age group.
Siesta time (1:00-3:00) - pedestrian count and questionnaire data affected as there were fewer people to collect data from. Pedestrian counts may have been significantly lower owing to the fact that there weren’t as many people around. Shops in the area closed up, particularly in Ibiza Town. This limited the number of potential people to question. To improve - increase the number of times we carry out the count (think about staying in evening to show variation in San Antonio).
Time of year - may have missed families as they tend to travel in the summer holidays however we went in mid-September. As it was closing weekend we saw predominantly young adults who were travelling for the clubbing scene, although our data at this time still may not be representative of the peak summer months.To improve - sample in peak summer months.
Land use mapping – people did not follow the criteria in the same way. Results may be unreliable as some people may have categorised one shop as retail but others as a gift shop, confusing the data. To improve - decide how to categorise the land use prior to the fieldwork. Have set definitions or a trial street so people are agreed upon how to classify the land use.
Bipolar survey – opinion-based, qualitative data. Large variation in the results we collected since some people found places more appealing than others, e.g. there is a difference of 23 between the highest and lowest bipolar scores awarded to site 3 in San Antonio. In order to improve our results we have calculated a mean from all the data we’ve collected. We could further improve this by removing any outliers.
Questionnaire – people were reluctant to participate in the survey. Bias was introduced since we could only survey English speakers so our results will not be representative. We were also more confident going up to elderly people to ask questions so this age group will be over-represented. To improve - we could have translated our questions prior to going out so that we could survey a more representative sample.
Pedestrian count – disputes over counting pedestrians twice. Some groups counted the same pedestrian twice if they walked past us more than once however others did not, this led to inaccuracies in our results. To improve - we should have resolved this prior to going out.
Our fieldwork aimed to investigate whether the null hypothesis ‘there is no variation in the type of tourism and its impact in the contrasting resorts or Ibiza Town and San Antonio’ is viable. Analysis of data generated from both primary and secondary sources illustrated that there are variations in the type of tourism in the resorts of San Antonio and Ibiza Town, and that these have led to various social, economic and environmental impacts, as predicted by the Butler Model, resulting in the development of government policy such as ‘A Taste of Spain.’
From conducting our pedestrian count we are able to conclude that there was an overall a higher number of pedestrians in Ibiza Town than in San Antonio. A distance decay effect was also evident in Ibiza Town as we moved away from the port, with figures of 142 and 102 in site 1 and 5 respectively which were on the main street by the port. This is approximately double the 50 pedestrians recorded at site 4, which was also the furthest site we recorded from the port in this location. We could therefore reject our null hypothesis. Due to the sampling sites being in such close proximity in San Antonio, no distance decay was seen.
We would have predicted that the number of pedestrians would have increased in San Antonio at night. However we were unable to sample at this time due to safety issues, thus our results do not reflect the full picture.
There is a clear contrast in bi-polar scores between Ibiza Town and San Antonio, as the mean bi-polar score for Ibiza Town is 3.4 and -7.6 for San Antonio. This difference of 11 therefore disproves our null hypothesis as it is suggested from the bi-polar scores that Ibiza Town is a more aesthetically pleasing location than San Antonio.
Ibiza Town attracts a different type of tourists than San Antonio, primarily older economic dependents and families due to the area being more cultural. The ‘Taste of Spain’ policy was implemented by the government to promote the culture of Ibiza Town, and help shed the island’s image as a ‘party island’ in order to attract high-end tourism. This is an example of rejuvenation within Ibiza Town.
Conversely, the tourists in San Antonio are mainly young economically active, generally between the ages of 16 to 30 years old; they account for 43% of the total tourist population in the area. Our land use surveys show that the most dominant type of land use in San Antonio is bars and pubs, and other land uses such as fast food (e.g. Subway) and retail are located in the area. These types of services attract younger tourists. San Antonio is currently in the state of stagnation where few new tourists arrive to the area and there is a local opposition to the type of tourism. San Antonio has not yet entered the state of rejuvenation where inward investment helps to re-establish the image and aid the development of both cultural and tourist aspects of the area.
The questionnaire showed that most tourists come from England in both sites, with a percentage of 36% in Ibiza Town and 43% in San Antonio, suggesting little variation. However, this does support the core-periphery model which suggests that tourists from the more affluent core will travel to the periphery to obtain a better deal.
However, there were variations in that there were more tourists from out of the continent in Ibiza Town than San Antonio, with some people we quentioned from Brazil, Australia and the USA, whereas 100% of those questioned in San Antonio were from Europe.
We also investigated the tourists’ length of stay in Ibiza which also showed little variation. For example, 23% stayed for 0-4 days in Ibiza Town compared to 21% in San Antonio. This similarity remained constant through all categories.
We also considered the ages of tourists in the two locations. We found that there are differences in age categories in the different locations which are clearly shown by our bar charts. We found out that 43% of tourists visiting San Antonio were in the 16-30 age category, whilst 66% of tourists in Ibiza Town were 31-60 years old. This suggests that San Antonio attracts younger people due to its reputable night life, whereas Ibiza Town is renowned for its cultural heritage and therefore attracts a more mature or family based demographic.
The land use transect we conducted showed more vertical land use in San Antonio. The types of shops were predominantly pubs, clubs and fast food store such as KFC found in site 4 and subway in site 3. This links to the theory of a Taste of Spain as it illustrates an area dominated by the clubbing scene. In contrast to site 4, site 5 was typically more retail club stores, which again links to the ideas in A Taste of Spain. There were also many souvenir shops here as well as at the bottom of site 4. This links to the Butler Model as it is perhaps the reason why Ibiza has gone into a state of stagnation.
After carrying out statistical tests on our data, we can reject our null hypothesis and conclude that there are variations in type of tourism and its impact in the contrasting resorts of San Antonio and Ibiza Town.
After carrying out our data collection, we have noted certain ways in which we could have improved in terms of removing bias, obtaining more representative data and gathering the data quicker and more accurately.
Disputes over counting a pedestrian twice were not settled prior to doing the fieldwork, reducing the accuracy of our data. Furthermore, if we had taken a mean value over several days we could have accounted for diurnal variations and our data would be more representative however as we were only there for a short period of time, this was not possible. There were also temporal issues in that the times at which we carried out the pedestrian counts were different at Ibiza Town and San Antonio which leads to inaccuracies.
Since this involved using our own judgement to rate the surroundings, the results were subjective. As a result this could limit the validity and reliability of our results due to different people rating each site according to personal opinion. Specific conditions should have been established prior to conducting the bipolar study to prevent high levels of bias. This would include looking at the criteria beforehand to discuss them as a group and establish what we would look for in order to give it a specific score. Additionally a sample or practice bipolar study could have been conducted as a group in order to demonstrate how to correctly use the criteria.
Originally we wanted to use random sampling to select the people we would question however this initial method was soon discarded owing to a lack of participation, and we were forced to ask anyone who was willing to help. Language barriers posed difficulties, meaning those speaking a foreign language were less represented in our data. In addition, the male population was sampled more due to a willingness to participate, particularly in San Antonio. Similarly, we were more confident in approaching elderly people so our data may over-represent this age group. Finally, safety issues had to be considered when we were approaching people, particularly in San Antonio where we came across many people who were under the influence of drugs and alcohol.
Land use transect:
To carry out our land use survey we selected an appropriate transect at each of the sites and then drew it according to the relevant height and width of the buildings with one square showing one storey and 2 metres across (about one average shop window). The process was very tedious due to the time it took to draw out each of the transects whilst also ensuring they were accurate. We could have taken photographs of each of the sites and taken them back home with us and drawn them out at a later date. In addition, when labelling the different shops according to a key, the categories were limited and there wasn’t always a suitable one, e.g. nothing for clubs in San Antonio. To prevent this, the categories should have been decided prior to carrying out the fieldwork in order to ensure the appropriate ones were being used, preventing inaccuracies in our data collection. Finally, it was difficult to determine how many squares to use when drawing out the transects, leading to some people drawing a shop front as two squares when others would only use one square, resulting in skewed data. Actual measurements should have been used (e.g. using a tape measure).
We then used the transects to complete a land use tally, which was done by adding up all of the squares under each of the categories (e.g. the number of squares labelled pub). However there were problems with the method in that some would count a shop that was 2 squares across as one shop whereas others would tally it as 2. This should have been discussed before and a joint decision reached.