Issues of reliability of validity of IMPLAN

Comments

49 comments

  • Avatar
    IMPLAN Forum
    Hello Kei, There is always a struggle with every survey to determine how much detail to include verses how much detail your respondents will be willing to fill out. If you feel 27 questions are too many, than you might choose to reduce the number. You certainly do not have to separate out the types of drinks. That was just an option if you had that kind of detail. It is completely fine to split these drinks into 2 categories 413 and 324. As regards seaweed, that could be a question that you choose to roll into a grocery question as 324 would be an appropriate Sector for any product purchased at a grocery store. However if you want to have that commodity detail, you can keep it out and Margin Sector 61. This is a personal choice for your survey. The difference between the methods we described yesterday, is that with the first method, which uses Sectors 320-330, Sectors that begin with the phrase "Retail Trade", you will only see the impacts of operating the Retail stores, which is a small portion of the value of the total sales (typically 20-30%). The other methodology allows you to estimate the impacts along the value-chain (producer, transport, wholesale and retail), this technically provides a more detailed estimation of the production process. However in you case, Malaysian processes may be different from those of Hawaii, and you have expressed numerous concerns about overestimation of impacts, so the Retail Sectors 320-330 may be the best estimation for you based on some of your concerns. Hope this helps. --Implan Support Staff
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    Thank you for the helpful suggestion. I guess I gotta roll the seaweed into grocery question by giving example Grocery (ie. seaweed) because I don’t want tourist to put seaweed under seafood category. Dried fish and seafood are commonly purchased by domestic tourists for own consumption. How about bakery & fruit? should I categorize them into grocery category? I will have problem with the package tour tourist, where the package covers accommodation, full board meal, transport, and others. I talked to a small lodging operator, he told me that a one night price (i.e US100) is set based on room 40% (US40), 50% full board meals (US50) because it is costly to transport the meal from mainland to island. However, that was just an answer given by one small operator. I am concern those big operator as they are not willing to tell the detail and make me hard to make estimation.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hello Kei, Again this is a personal choice, but you certainly can roll bakery items in a split between 324 and 413. Just remember that if it's purchased at a restaurant, like the meal described below, that will be 413 not 324. 413 will purchase all the food ingredients needed to make meals, so you don't have to worry about tracking individual items that go into restaurants and bars. Again with 324, you can get some additional information and do a more detailed study, but there is nothing wrong with just saying how much did you spend at the grocery vs. at restaurants. One other thought on this, there are both Alcoholic Beverage and Food and Non-alcoholic beverage spending patterns in the InvestmentPatterns_for_BEAIndustry_MIG_Lib440.impbdest that you could use to proxy purchase of alcoholic beverages at retail stores and grocery sales. You could use these as way to approximate more detail than your survey allows due to reason of practicality as you mentioned earlier. If choose to use these spending patterns you might consider modifying them based on collected data from your survey or 'common sense' observation to make them a little more specialized to your region. For example, you might weight fish and seafood products higher and eliminate poultry or beef if you know those are not common or not present in your local grocers. If you are interested in using these spending patterns they are found at Activity Options> Import> From Another Model. Direct the file structure back to the IMPLAN User Data folder and then select Utilities folder. Please let us know if you have any additional questions.
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    Hi, I am finalizing the expenditure categories as I am going to collect the data next week. I have a question, should i use dried fish and seafood as one category instead of putting them into grocery category? Dried fish and seafood are normally purchased at wet seafood market and dried seafood market.Thank you.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hi Kei, If you want to get the extra data and you have a Sector in mind to map it to, you can certainly get this extra piece of information. If you don't, but you want to be able to report information on these types of purchases separately for your report, you can certainly record the data and report it, but you would roll-up the values of these purchases into the grocery Sector for analysis. Please let us know if you have any additional questions.
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    Thank for the explanation. There are massage services, as well as spa service offered by hotel. So, i have one category for massage service, IC 419. However, massage service is not popular in my study area but spa are popular in some big resorts as there are many chinese tourists. I afraid my respondents feel boring when they see my questionnaire as too many categories >_< Well, I have no choice as i need to get more accurate information.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    You are most welcome. That is always a challenge, as we have discussed. You can certainly try a higher level survey and if you are unable to get successfully completed responses, roll up categories to try to help your survey recipients to complete the questions. Please let us know if you have any additional questions.
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    Happy New Year 2014! I just came back from my data collection, i did collect the expenditure on air transport. As mentioned earlier, i will only capture the expenditure within my study area. In order to travel to my study site, one can either travel via air or land. However, some respondents gave me the return fare instead of one way fare. Should I record one way or return fare?Or should I divide the fare by two? If the respondent travel with friend and they were sharing the room, should i divide the expenditure by two (twin sharing)? The expenditure is based on per person per trip, if one pay for his family for F&B (e.g 4 people), should i divide by 4? Thank you
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hi Kei, If you are running your study on a per visitor basis, you will want to divide costs by the number of individuals in a group. For transport fare, typically only half of that goes to the destination and the other half goes to the point of the origin. If this is true for transport other than air in your region than you would only want to count the value spent in your region (destination on arrival/point of origin for departure) typically this can be estimated by taking half of the total travel costs. Hope this helps! Please let us know if you have any additional questions.
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    I am also facing a problem on package tour, as indicated in the earlier post. Some respondents reported a lump sum payment (inclusive of accommodation, food, airport transfer and boat transfer to island). I know some of the break down cost for package tour from a budgeted tour operator. the breakdown cost 50% for fnb (including 3 meals) and lodging 40%, 10% other. some packages exclude airport transfer ($30-50), boat transfer ($100). But, I fail to get the break down cost for luxury accommodation (confidential) which maybe different from budgeted one. Can I make assumption the breakdown cost are the same?
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hi Kei, Unless you can find a better way to break it down, then that is probably the best method you can use to split those costs. You might also see if any previous studies have been done by others on your region or similar regions that maybe were able to get alternative breakdowns or estimates. Do you have a tourism office that could perhaps help? If you can't find any additional information our recommendation would be to use what you have and then just state as part of your study assumptions that you had to use the same breakout for all packages because no additional data on this breakdown could be found. Hope this helps!
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    I have read many studies but I could not find any. Often, the reported expenditure is quite straightforward with no package cost involves. I do not have a tourism office that could help. Resort operator is the only one could assist me; however that's confidential (big operator) for them to tell the breakdowns or estimates. The only thing i can do is go through their website and see what package they are offering and make assumption. The package is based on per person basis, including the land and sea transfer. That could be an alternative for me, but I just worry about the validity and reliability claims. For example a 3 days 2 nights package offered by luxury resort cost 1500, it doesn’t make sense visitor spend 50% (750 for 2 days) on food and beverage (3 meals per day).
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hi Kei, Unfortunately in these examples there are no absolutely correct answers. That is why we suggest that you choose what seems to be the best assumption from what is available to you. The reason that we recommend presenting these assumptions in your report documentation, is that if someone disagrees with this assumption then they know what you have done based on the data that was available to you. If they can provide you with additional data, you can then reexamine the impact based on the new information.
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    Thank you for your suggestion. Should I report one way or return expenditure on land transport (taxi)?
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hello Kei, We apologize, but we aren't quite sure what you are asking here. Could you please provide us some additional information in this regard. Thanks!
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    For land transport, visitors can either choose by taxi or by bus, which is outside the study boundary. Some respondents reported return fare and some report one way fare. My question is should i report one way fare (as what i need to report for the air transport) or return fare (Of course the payment for bus or taxi will be made within the study boundary)?
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hi Kei, If all the transport companies are outside of your Study Area region you won't want to count those dollars anyway. The key will lie in whether the company is local to your region. Even if the payment is made in the region, if the company is outside the region, that money can be considered as leaking. If the company is inside the region or has representation (like a satellite office) in your region than you can assume half the total fare as local. Hopefully this help and we apologize for the delay in the response.
    0
    Comment actions Permalink
  • Avatar
    Kei Wei
    I was collecting on-site survey on expenditure. If tourist were approached on day three, they reported the length of day of 5 days. Should I report the expenditure until day three or should I forecast their expenditure until day five? If respondents did not report their expenditure on air transport, should i ignore or predict price based on the basic fare (some might get promo fare)? Thank you.
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Forum
    Hello Kei, That is really up to you, either way we would recommend stating in your report what you choose. You certainly can try to take the data that you have and create an average daily expenditures pattern for your visitors in that grouping and then use this to estimate their expenditures fro the remaining two days. If you do so you would just want to state that you estimated the last two days of their trip based on average daily expenditures reported from the first three days. If you are not comfortable making that estimate you should report that you know that the average guest in that group remained five days, but you only are reporting three days of their expenditures because that is all the data that was collected. Thus the impact you are reporting will be somewhat conservative. The same principles apply to airfare, you could use what the average reported expenditures on airfare (remembering that you only want to account for half the value since only half the trip was at your local airport- and assuming that the airport is within you Study Area region) as an estimate of what airfare would be for those who did not report it. You would just want to state that you estimated based on average airfare rates of others surveyed, or in general depending on your data, or you chose not to report it because it wasn't local or because you are not comfortable making an estimation.
    0
    Comment actions Permalink

Please sign in to leave a comment.