Directorate
The Intellectual Property Policy Directorate (IPPD) has the primary responsibility in Industry Canada for the development and coordination of marketplace framework policies related to intellectual property policy including copyright laws. IPPD's role is to advise the Minister of Industry and senior departmental officials on intellectual property matters except for patents (which fall under the responsibility of the Patent Policy Directorate (PPD)), and together with other interested federal departments, to develop specific policy proposals and legislative initiatives in these areas.
Along with the Copyright Policy Branch at Canadian Heritage, IPPD develops significant policies on copyright matters. IPPD also works closely with the Department of Foreign Affairs and International Trade, and provides key Canadian representation on intellectual property issues internationally.
The Copyright Act (the Act) encourages the creation and dissemination of original material by granting to creators the right to derive income from and control certain uses of their material. However, to further other economic and social objectives and to ensure widespread dissemination and reasonable access to the ideas embodied in the material, the Act also limits the scope and duration of these rights. Given the Act's importance as part of the overall economic framework, and as a means for promoting innovation and R&D through knowledge creation and knowledge dissemination, it plays a central role in Industry Canada's (IC's) mandate to enhance Canadian participation in the knowledge economy.
Rationale
Industry Canada has commissioned a Decima Research survey to collect data on the music purchasing and Internet peer-to-peer (P2P) file sharing activities of Canadians. The objective of the survey was to fill a major data gap in Canada on the issue of music file sharing activities of Canadians.
While there is growing literature on file sharing via P2P and its impact on pre-recorded music sales, none of the existing studies has analyzed Canadian data, let alone focused on the Canadian case. One of the reasons for this is the absence of reliable Canadian data.
Anticipated Outcomes
The data will be used in a statistical analysis intended to quantify the relationship between music consumption and peer-to-peer (P2P) file sharing and other socio-economic variables, such as age, gender and income.
The results will be taken into consideration, along with other information and analysis, and will inform Industry Canada's policy development work.
Research Information
Quantitative:
Telephone interviews were conducted with 2 100 randomly selected Canadians (15 years and older) across the country between April and June 2006. The questionnaire was developed by Industry Canada in collaboration with an outside consultant and Decima Research. The questionnaire was pre-tested in both official languages.
Telephone survey of 2,100 Canadians – 1,000 who download music files over the Internet and 1,100 who do not.
Research Firm: Decima Research
Contract Number: U5500-050472
Contract Issued by: Public Works and Government Services Canada
Contract Value: $112,356.86
- Music File Sharing Study 2006 — Methodology Report, August 31, 2006
- Music File Sharing Survey Questionnaire for Canada
- Music File Sharing Study 2006, August 2006
- Music File Sharing Study 2006—Final Data
Music File Sharing Study 2006—Methodology Report—August 31, 2006
Registration number: POR-335-05
Contract number: U5500-050472
Proprietary Warning
The information contained herein is proprietary to Industry Canada and may not be used, reproduced or disclosed to others except as specifically permitted in writing by the originator of the information. The recipient of this information, by its retention and use, agrees to protect the same and the information contained therein from loss, theft or compromise. Any material or information provided by Industry Canada and all data collected by Decima will be treated as confidential by Decima and will be stored securely while on Decima's premise (adhering to industry standards and applicable laws).
Decima Research
Toronto
2345 Yonge Street
Suite 405
Toronto, Ontario M4P 2E5
Telephone: 416-962-2013
Fax: 416-962-0505
Ottawa
160 Elgin Street
Suite 1820
Ottawa, Ontario K2P 2P7
Telephone: 613-230-2200
Fax: 613-230-9048
Montreal
1080 Beaver Hall Hill
Suite 400
Montreal, Quebec H2Z 1S8
Telephone: 514-288-0037
Fax: 514-288-0138
Decima Research Inc. is ISO 9001:2000 Certified
Table of Contents
IntroductionMethodology- Overview
- Sampling Distribution
- Research Material Design
- Pre-testing
- Sample Selection and Contact
- Survey Administration
- Sample Management
- Sample Weighting
- Follow-Up Qualitative Analysis (Post-Hoc)
Introduction
Background
The widespread phenomenon of music file sharing over the Internet began in 1999, with the creation of the Napster program. This was a software program that could be downloaded onto a personal computer from the Internet free of charge, allowing users to exchange mp3 format audio files. Napster was the very first peer-to-peer (P2P) computer software programs, requiring a central server to connect users. The current generation of P2P networks does not operate through a central server, as each user establishes contact with others directly.
There is a growing literature on music file sharing and its impact on prerecorded music sales. However, none of the existing studies have analyzed Canadian data, let alone focused on the Canadian case. The primary reason for this is the absence of Canadian data.
An effective means of collecting data on the variables of interest is through a survey of Canadians. Ultimately, this survey will help Industry Canada derive accurate and statistically reliable metrics around the following:
- Counts of the number of files shared over the selected time period;
- Purchases of pre-recorded music products (PMPs) over the selected time period;
- Number of Canadians who purchase PMPs; and
- Number of Canadians who engage in music file sharing activity.
Purpose
The purpose of this document is to detail the methodological considerations used in collecting this data. It clearly outlines assumptions used, procedures followed, and why certain decisions were made in mapping this data.
Methodology
Overview
This survey was conducted using a traditional telephone survey data collection methodology (CATI — Computer Assisted Telephone Interviewing). Specific dimensions of the research included:
- 2,100 telephone interviews with Canadian households over the period spanning April 19th through May 20th;
- National representation with special consideration by age, gender and region;
- Interviews were segmented based on whether or not the respondent had downloaded free music from a P2P (peer-to-peer) network during 2005;
- Quota-based, weighted distribution;
- Average survey length of 10 minutes;
- Interviews were conducted with respondents 15 years of age and older;
- Respondents were informed that the survey was being conducted by Decima Research on behalf of Industry Canada (i.e. sponsorship revealed study).
Sampling Distribution
A stratified, quota-based sampling approach was used, since this generates substantive estimates across and within specific key segments of interest, which in turn permits extrapolation to the broader population with greater confidence. Previous research suggests that music purchasing and file sharing behaviours tend to differ according to a person's mother tongue, gender and age. Quotas were therefore set by controlling for age, gender and region, where region is used as a proxy for mother tongue — that is, Quebec is treated as Francophone. Furthermore, given the importance of understanding behaviours among both music downloaders and nondownloaders, quotas were introduced to control for this as well.
The grid below summarizes the final sample distribution across the quota cells. Industry Canada requested 1,000 responses for both downloader and non-downloader segments and approximately 100 responses per quota cell in order to meet desired standards of statistical reliability.
Region | Age/Gender | Downloaders | Non-downloaders | Total |
---|---|---|---|---|
Quebec | Male | 160 | 125 | 285 |
Male 25+ | 90 | 125 | 215 | |
Female | 160 | 125 | 285 | |
Female 25+ | 90 | 131 | 221 | |
Total Quebec | 500 | 506 | 1006 | |
Rest of Canada | Male | 160 | 132 | 292 |
Male 25+ | 90 | 153 | 243 | |
Female | 160 | 133 | 293 | |
Female 25+ | 90 | 176 | 266 | |
Total Rest of Canada | 500 | 594 | 1094 | |
Total | 1000 | 1100 | 2100 |
The maximum margins of error associated with the above segmentation are as follows:
Population Segment | Sample size | Margin of error |
---|---|---|
Downloaders or non-downloaders | N=~1000 | +/- 3.1% |
Quebec or Rest of Canada | N=~1000 | +/- 3.1% |
Downloaders or non-downloaders in each region | N=~500 | +/- 4.4% |
Males or females under 25 in each region | N=~285 | +/- 5.8% |
Males or females over 24 in each region | N=~215 | +/- 6.7% |
Non-downloaders within each age, region and gender segment | N=~125 | +/- 8.8% |
Downloaders within each age and region segment, under 25 | N=160 | +/- 7.8% |
Downloaders within each age and region segment, over 24 | N=90 | +/- 10.3% |
Research Material Design
The questions for the survey were designed in conjunction with the Industry Canada team, using the following iterative process:
- Industry Canada initially provided Decima Research with a draft questionnaire for review. Decima then provided recommendations regarding language used, survey flow, skip patterns, programming and interviewing instructions. An updated draft version was subsequently returned to Industry Canada, who then made modifications accordingly. Once all parties were satisfied that the questionnaire was ready to be pre-tested, the survey was programmed using CATI software.
Pre-testing
A pre-test was performed with 23 English and 23 French respondents. The general purpose of this pre-test was to evaluate respondents' comprehension of the questions, the completeness of the response categories, and overall response to the survey (…is there anything that might make respondents uncomfortable or angry? …are certain questions generating an abnormally high amount of non-response?). Further, two versions of the draft questionnaire were tested — "Version A", which asked respondents toestimate the total number of items purchased over a specified period (e.g. number of CDs purchased in 2005) and "Version B", which asked respondents to specify the total amount spent (e.g. amount spent on CDs in 2005). Respondents were randomly selected to receive either "Version A" or "Version B".
The distribution of surveys by version and by language is as follows:
Version A | Version B | |
---|---|---|
English | 10 | 13 |
French | 12 | 11 |
Total | 22 | 24 |
Following the pre-test, Decima provided a summary of the results and recommendations for modification to the questionnaire. Among other issues, the pre-test revealed that ‘Version A' was easier for respondents to answer. Based on the pre-test results, a number of modifications were made to the questionnaire.
Sample Selection and Contact
Sample Selection
Decima created the sampling frame using Canada Survey Sampler (CSS). CSS is a proprietary selection engine specifically designed to generate a random sample of telephone numbers to be dialed. The CSS maintains a comprehensive list of all populated exchanges across Canada, and is updated on a regular basis.
The CSS works by randomly generating 4-digit suffixes for these exchanges. These suffixes are generated in proportion to the percent population of the individual exchanges (i.e. a 90% populated exchange would experience twice as many 'hits' as a 45% populated exchange). As each suffix is generated, it is compared to the database of existing known phone numbers. If it matches a listed phone number, it is placed into the 'valid number' file. If not, it is placed in the 'orphan' file. Decima uses the valid number file as its primary calling list. This list is then supplemented with numbers from the orphan list. As with the random generation above, numbers are chosen from the orphan list in proportion to the percent population of the exchanges.
The supplementing of the sample file with the orphan file is intended to ensure that our sample scheme accurately emulates what is know as a 'Waksberg' RDD design, but is more efficient because it more effectively includes unlisted and new telephone numbers while significantly reducing the number of ineligible "not in service", fax and cell phone numbers that would normally be encountered with conventional RDD methods.
Respondent Selection
Because interviews tend to be completed by an adult in the household (notably those 20 years of age or older), once a household was contacted, the respondent was asked whether there was anyone in the household between the ages of 15 and 19 who could complete the survey, to ensure sufficient representation of this younger cohort within the youth segment.
Survey Administration
All interviewing was conducted through Decima's field division, Opinion Search, which operates one of the largest and most sophisticated research data collection operations in Canada. From its offices in Ottawa, Toronto and Montreal, Decima utilizes 325 state-of-the-art computer-assisted-telephone-interviewing (CATI) stations, which are fully networked and utilize sophisticated VOXCO technology that is among the most powerful on the market today.
Interviewing was conducted in English and French, based on the expressed preference of the respondent. Up to eight call-backs were made to each eligible or potentially eligible household during the interview period before classifying it as "not available." Fifteen percent of all calls were monitored by the supervisory staff. Frequency checks were administered throughout the data collection period to ensure that all questions were correctly programmed.
Sample Management
Sample was allocated in small segments and exhausted before new sample was substituted. All contact dispositions were tracked and recorded in the sample database. In many instances, the system automatically allocated sample records in response to recorded disposition. For example, the system automatically rescheduled callbacks for "no answers" and "busy" dispositions in a pre-set pattern to compensate for potential non-response on the occasions when individuals were not at home or their lines busy. For example, unanswered calls in the early evening could be scheduled for another day later in the evening; a weekend call could be scheduled for weekday evening and vice versa. For busy line signals, the schedule was created to re-attempt the record within a set time frame that same day. This entire process is consistent with the PWGSC standing offer for public opinion research.
Sample Weighting
Decima uses well-defined procedures for calculating weighting factors, based on established methodological standards and extensive experience in sample weighting on over literally hundreds of projects. This procedure involves calculating the actual population within each quota segment (region, age group, gender and downloader status), and the true proportion of the sample that would fall into each segment if the survey were conducted in strictly a random basis (based on the most current population statistics available from Statistics Canada). Into this number is divided the actual segment sub-sample to produce a weighting factor that is then used to "weight" the data for that segment. While there are various ways of accomplishing this task, this procedure is the most straightforward and effective.
In this instance, the actual population of downloaders vs. non-downloaders was not known (or was not considered reliable). Thus, these proportions were estimated based on how sampling "naturally" occurred at the beginning of the survey period (i.e. up to the last day before any quota cells were closed). Details of this process are outlined in two steps.
Step 1. First, the true population proportions are determined by taking into account age, gender, region and downloader status, since these are the variables on which the quotas were based.
Using Statistics Canada Census 2001 data, we know the population proportions by age, gender and region, as seen below:
Canada | Quebec | Atlantic | Ontario | West | |
---|---|---|---|---|---|
Male Under 25 | 2,034,430 | 482,000 | 154,505 | 754,555 | 643,355 |
Male 25+ | 9,741,425 | 2,391,760 | 747,850 | 3,677,835 | 2,924,095 |
Female Under 25 | 1,974,705 | 467,480 | 153,125 | 733,275 | 620,800 |
Female 25+ | 10,530,965 | 2,604,635 | 819,730 | 4,011,665 | 3,094,940 |
For the downloader status, we look at the quota cells in the earlier days of the study before any of the quota cells were closed, since this is our best indicator of how the proportions naturally fall out when random calling is used. This distribution is seen below, and is based on a sample size of n=691 overall:
Type | Quota | Completed | Distribution |
---|---|---|---|
320 | 50 | 7.2% | |
25+, Downloader, Male | 180 | 64 | 9.3% |
250 | 42 | 6.1% | |
25+, Non-Downloader, Male | 250 | 176 | 25.5% |
320 | 31 | 4.5% | |
25+, Downloader, Female | 180 | 53 | 7.7% |
250 | 51 | 7.4% | |
25+, Non-Downloader, Female | 250 | 224 | 32.4% |
We assumed that the distribution within each region would follow the same pattern (because the overall sample size was relatively low, it was not feasible to look at the natural distribution within each region). Based on the national distribution, we calculated the population distribution within each region. Calculations for Quebec are provided in the table below.
Quota (Quebec) | Distribution | Population |
---|---|---|
7% | 43,0237 Footnote * | |
25+, Downloader, Male | 9% | 550,703 |
6% | 361,399 | |
25+, Non-Downloader, Male | 25% | 1,514,434 |
4% | 266,747 | |
25+, Downloader, Female | 8% | 456,051 |
7% | 438,842 | |
25+, Non-Downloader, Female | 32% | 1,927,462 |
For the final proportions, we needed to make them representative of the entire targeted Canadian population.
Quota (Quebec) | Population | Canada Population | Proportion within Canada |
---|---|---|---|
430,237 | 24,281,525 | 2% | |
25+, Downloader, Male | 550,703 | 24,281,525 | 2% |
361,399 | 24,281,525 | 1% | |
25+, Non-Downloader, Male | 1,514,434 | 24,281,525 | 6% |
266,747 | 24,281,525 | 1% | |
25+, Downloader, Female | 456,051 | 24,281,525 | 2% |
438,842 | 24,281,525 | 2% | |
25+, Non-Downloader, Female | 1,927,462 | 24,281,525 | 8% |
Step 2. Finally, we looked at our true sample numbers for each group within each province. For example, we collected 160 Quotas for this study were based on a set of questions asked at the beginning of the survey to categorize the respondents into their respective quota cells. In order to keep survey costs at a minimum, calls made to individuals that fall into a "filled" quota cell are terminated as we continue to target the harder-to-reach quota cells. A general set of questions related to music behaviour are asked upfront, which are then used for skip logic so that, for example, a respondent who did not purchase any CDs in 2005 will not be asked questions related to CD purchasing behaviour later on in the survey. Similarly, our downloader screening question was asked in this early section of our survey, where we ask the respondent, "Did you download music from P2P (person-to-person) networks in 2005?", where the respondent had the option of giving a "yes" or "no" answer. Those who answered "yes" were asked questions in "Section 4: The Effects of Unauthorized mp3 Downloading". Question 4.1b of the survey asks, "How many free music tracks did you download from P2P networks in an average month in 2005 " Normally, it is expected that 1%-3% of respondents arrive at this section, and then give a "don't know" or non-behavioural response (i.e. zero downloads). In this instance, 246 of our 1,000 quota gave a "zero" response or answered "don't know". Given the magnitude of this proportion of respondents, additional analysis was warranted to better understand true downloading behaviour. Post-hoc analyses were conducted to determine if these individuals should be categorized as downloaders or non-downloaders. Respondents who answered "don't know" are deemed to be downloaders for two main reasons: The more complex group to profile were those who answered "zero" to Q4.1b. We decided to go back into field and target those respondents who answered "zero" to Q4.1b. The following question was asked: "In the survey, a number of Canadians indicated that during 2005 they downloaded free music from P2P file sharing networks, like Kazaa, LimeWire or Archambeault. But then when we asked them to specify the number of free music tracks they downloaded in an average month during 2005, many individuals answered 'zero'. So we're calling some Canadians back to try and understand this disconnect. We have identified a variety of possible reasons. Let me know if any of these apply to you. Is it because…" In this qualitative exercise, twenty-five respondents were targeted over the course of two evenings, and in answering this question, the intention was to help us understand how these individuals should be categorized for the analysis – that is, as downloaders or non-downloaders. The reasons for the 'zero' responses can be divided into three main areas: they download sporadically, and not on a regular, monthly basis; they used to download from P2P networks but do not any longer and they do not download from P2P networks at all. Another less common reason is that they download few songs per year (i.e. less than one per month). Based on these qualitative findings, we decided to assign these respondents a "downloader" status as well, with the subtle distinction being that they are not "regular" downloaders. The chart below identifies the number of total attempted calls and the nature of these calls. Decima Final Questionnaire Good [morning/evening], my name is _______ and I'm calling from Decima Research. We are conducting a survey on behalf of the Government of Canada's Department of Industry. This is a national survey to obtain data on Canadians' music purchasing and listening habits. Your telephone number was randomly selected and all responses will be kept strictly confidential and analysed in the aggregate. Would you have time to help us out this [morning/evening]? 1.1 Is there anyone in your household currently between the ages of 15 and 19 inclusive? We are looking to complete surveys with young Canadians to gain their unique insights — would it be possible to complete the survey with this individual? 1.2 Could you tell me whether you are under 25 years old or 25 years of age or older? 24 years old or younger 1.3 Since this survey is going to revolve around music habits of Canadians, I'd like to ask you about the ways in which you acquired music during 2005. If you are in any way uncertain of the meaning of any terminology referred to in this survey, please do not hesitate to ask me for clarification. Please indicate whether you obtain music in the following ways: [Skip to 2.2 if did not purchase CD albums in 2005 (1.3a≠1)] 2.1 (a) In 2005, how many music CDs did you buy for your personal use? (No categories provided) _______________(#) Respondents to be first asked 2.1a, and if unable to provide answer, ask a2.1b. Should not be an answer for both questions. Intention is to minimize "Don't know" responses. 2.1 (b) Would you say that the number of music CDs that you purchased for your personal use in 2005 was… [Read list] 2.2 (Ask if 2.1a≠0 or 2.1b≠0) Did you purchase more, less or about the same number of music CDs for your personal use in 2005, as compared to 2004? (Ask if 2.1a=0 or 2.1b=0 or 1.3≠1) Did you purchase less or about the same number of music CDs for your personal use in 2005, as compared to 2004? 2.3 (Ask if 2.2=1 or 2) How many [More/Fewer] music CDs did you purchase for your personal use in 2005, as compared to 2004? ______________(#) [Skip 2.4 if (2.1a=0 or 2.1b=0) or (1.3a≠1)] 2.4 What was the price per CD purchased during 2005? $___________ [Skip 2.5 if (1.3a≠1 or 2.1a=0) and (2.2=3)] 2.5 (a) Did the average price of music CDs increase, decrease or stay the same in 2005, as compared to 2004? (b) By how much did the price of music CDs [Increase/Decrease] in 2005, as compared to 2004? ______________($) [Skip 2.6 if do not purchase CDs (Q1.3a≠1)] 2.6 (a) In 2005, how many music CDs did you buy as gifts for family or friends? (No categories provided) ________(#) Respondents to be first asked 2.6a, and if unable to provide answer, ask a2.6b. 2.6 (b) Would you say that the number of music CDs that you purchased as gifts in 2005 was… Read list: [Skip 2.7 if (1.3f≠1)] 2.7 (a) In an average month during 2005, how many music CDs did you rip? [If necessary: That is, how many CDs did you copy onto your computer?] _________________(#) (b) In an average month during 2004, how many music CDs did you rip? _________________(#) [Skip 2.8 if (1.3b≠1)] 2.8 (a) In an average month during 2005, how many paid music tracks did you purchase from pay-sites such as iTunes or Archambault ? ______________(#) Respondents to be first asked 2.8a, and if unable to provide answer, ask 2.8b. 2.8 (b) In an average month during 2005, would you say that the number of paid music tracks you purchased from pay-sites was… [Read if necessary: Pay sites include iTunes, PureTracks, Archambault and Napster] Read list: 2.9 Did you purchase ["more" Not if 2.8a=0 or 2.8b=0 or 1.3b≠1], less or about the same number of paid music tracks in 2005, as compared to 2004? [Read if necessary: Pay sites include: iTunes, PureTracks, Archambault and Napster] 2.10 [Ask if 2.9=1 or 2] How many [More/Fewer] paid music tracks did you purchase in 2005, as compared to 2004? ______________(#) [Skip 2.11 if (2.8a=0 or 2.8b=0 or 1.3b≠1)] 2.11 Thinking of the paid music tracks that you purchased from pay sites in 2005: [Read if necessary: Pay sites include iTunes, PureTracks, Archambault and Napster] What was the price per track in 2005? $___________ [Skip 2.12 if (2.8a=0 or 2.8b=0) and (2.9=3)] 2.12 Did the price of paid music tracks increase, decrease or stay the same in 2005, as compared to 2004? [Read if necessary: Pay sites include iTunes, PureTracks, Archambault and Napster] 2.13 (Ask if 2.12=1 or 2) By how much did the price of paid music tracks [Increase/Decrease] in 2005, as compared to 2004? ______________($) 3.1 (a) Did you purchase any non-writable DVDs for your personal use during 2005? (b) [Ask if 3.1a=1] How many DVDs did you purchase during 2005? ____________(#) (c) [Ask if 3.1a=1] What was the average price per DVD? ___________($) 3.2 (a) Did you purchase any video games for your personal use during 2005? (b) [Ask if 3.2a=1] How many video games did you purchase during 2005? ____________(#) (c) [Ask if 3.2a=1] What was the average price per video game? ______________($) 3.3 (a) Did you go to the cinema during 2005? (b) [Ask if 3.3a=1] How many movies did you go to during 2005? ____________(#) (c) [Ask if 3.3a=1] What was the average price per movie ticket? _______________($) 3.4 (a) Did you attend any live concerts during 2005? (b) [Ask if 3.4a=1] How many live concerts did you attend during 2005? _________(#) (c) [Ask if 3.4a=1] What was the average concert ticket price? ____________($) [Skip to 4.9 if no P2P downloading (4.1bi=0 and 4.1bii=0)] 4.2 Please indicate how frequently you use each of the following for storing MP3 files that you acquire. Please use a scale from 1 to 10, where '1' means 'you never use this storage method' and '10' means 'you always use this storage method'. (a) Hard drive/ CPU 4.3 (skip if 4.1bi=0) Considering the songs that you downloaded from P2P networks during 2005: (a) What percentage did you purchase as paid music tracks from pay-sites? [Interviewer note: 0% means that they did not subsequently purchase any tracks; 100% means that they subsequently purchased all tracks] ______________(%) (b) What percentage did you purchase as part of a music CD? [Interviewer note: 0% means that they did not purchase any tracks as part of a CD; 100% means that they purchased all tracks as part of a CD] ______________(%) 4.4 (skip if 4.1bi=0) Considering the songs that you downloaded for free from P2P networks during 2005: (a) What percentage would you have purchased as paid music tracks if they were not available through P2P networks? [Interviewer note: 0% means that they would not purchase any tracks; 100% means that they would purchase all tracks] _______________(%) (b) What percentage would you have purchased as part of a music CD if they were not available through P2P networks? [Interviewer note: 0% means that they would not purchase any tracks; 100% means that they would purchase all tracks] _______________(%) 4.5 (skip if 4.1bi=0) What percentage of songs downloaded from P2P networks do you erase after listening to them only once or twice? [Interviewer note: 0% means that they keep all music that is downloaded; 100% means they erase all music that they downloaded] __________________ % 4.6 (a) (skip if 4.1bi=0) If the price of paid music tracks from sites like iTunes were reduced by half, do you believe you would download fewer or about the same number of free music tracks from P2P networks? 4.7 (b) By what percentage would you reduce your free downloads from P2P networks? [Interviewer note: 0% means that they would not reduce at all; 100% means they would completely stop downloading from P2P] ________________% 4.8 (skip if 4.1bi=0) Thinking of the music tracks that you downloaded from P2P networks during 2005, what percentage did you download because: 4.9 (a) Between 2004 and 2005, do you feel that the quality of music improved, declined or stayed the same? By "quality" I am referring only to musical content. (b) Over the course of 2005, do you feel that the quality of music improved, declined or stayed the same? [Repeat if necessary: By "quality" I am referring only to musical content] 4.10 How would you describe your interest in listening to music? Would you describe it as… 4.11 How would you describe your skill level in using the Internet? I have just a few final questions that will help us analyze the results… D1 Please stop me at the age category to which you belong: (Read and pause after each) D2 Please stop me at the highest level of schooling you have attained: (Read and pause after each) D3 Which of the following best describes you? Are you…? (Read and pause after each) Ask if D3=1 (students only) D4 Do you work… D5 Which of the following categories best reflects your household's total income before taxes in 2005. That means the total income earned by all those living in your home? D6 Compared to your total household income in 2004, did your total household income in 2005… D7 (If D6=1 or 2) In either percentages or dollars, how much did your total household income in 2005 [Increase/Decrease] from 2004? __________% or $____________ D8 code gender of respondent (Do not ask) D9 Language of survey (Recorded by CATI) Date: August 2006 Comparison Groups: BC/ DE/ FG/ HI/ JKLM/ NOPQ/ RSTU Comparison Groups: BC/ DE/ FG/ HI/ JKLM/ NOPQ/ RSTU Comparison Groups: BC/ DE/ FG/ HI/ JKLM/ NOPQ/ RSTU Comparison Groups: BC/ DE/ FG/ HI/ JKLM/ NOPQ/ RSTU Follow-Up Qualitative Analysis (Post-Hoc)
"Don't Know" Responses
"Zero" Responses
Appendix: Call Disposition Report
A (1-14) Total Attempted 44186 1 Not in service 4508 2 Fax 711 3 Invalid#/Wrong# 1,496 B (4-14) Total Eligible 37,471 4 Busy 273 5 Answering machine 6,327 6 No answer 4,846 7 Language barrier 855 8 Ill/Incapable 376 9 Eligible not available/Callback 2,032 C (10-14) Total Asked 22,762 10 Household/Company Refusal 6,516 11 Respondent Refusal 10,369 12 Qualified Termination 186 D (13-14) Co-operative Contact 5,691 13 Not Qualified 3,591 14 Completed Interview 2,100 Refusal Rate 75.00 (10+11+12) / C Response Rate 15.19 D (13-14) / B (4-14) Incidence * 38.90 [(14+12) / (13+14+12)]*100 [(CI+QualTM)/(NQ+CI+QualTM)]*100 Music File Sharing Survey Questionnaire for Canada
April 11, 2006
0. Entry question
1. Screening/Profiling
Nobody of this age 1 Continue with original respondent Speaking 2 Go to main survey [skip to 1.3] Gets the person 3 Repeat intro Refusal 4 Thank and terminate
25 years old or older Yes No a) Buy music CDs. 1 2 b) Buy music tracks from pay-sites like iTunes or Archambault 1 2 c) Download free music from promotional websites [If necessary: For example, myspace.com or a band's website] 1 2 d) Download free music from P2P file sharing networks, like Kazaa or LimeWire [If necessary: Other examples are eDonkey, BearShare and Gnutella] 1 2 e) Copy MP3s from friends. 1 2 f) Rip songs from CDs [If necessary: Copy CDs onto your computer] 1 2 g) Download music from peoples' private Internet websites [If necessary: Personal non-commercial websites whose content can be downloaded free of charge] 1 2 2. Music purchasing and pricing
None 0 1 to 2 1 3 to 5 2 6 to 10 3 11 to 15 4 16 to 20 5 21 to 30 6 31 or more 7 More 1 (Not if 2.1a=0 or 2.1b=0 or 1.3≠1) Less 2 Same 3 Increased 1 Decreased 2 Stayed the same 3 None 0 1 to 2 1 3 to 5 2 6 to 10 3 11 to 15 4 16 or more 5 Paid Music Downloads
[Read if necessary: Music track formats include MP3, AAC and WMA] None 0 1 to 5 1 6 to 10 2 11 to 20 3 21 to 50 4 51 or more 5 More 1 (Not if 2.8a=0 or 2.8=0 or 1.3b≠1) Less 2 Same 3 Increased 1 Decreased 2 Stayed the same 3 3. Substitute Entertainment Goods
Yes 1 No 2 Yes 1 No 2 Yes 1 No 2 Yes 1 No 2 4. The Effects of unauthorized MP3 downloading and Miscellaneous
Source Average #/month in 2005 (i) And how many per month in 2004? (ii) 4.1a. (Skip if 1.3c≠1) How many free music tracks did you download from promotional websites in an average month during 2005? (If necessary: For example, myspace.com or a band's website)? 4.1b. (Skip if 1.3d≠1) How many free music tracks did you download from P2P networks in an average month during 2005? (Read if necessary: Examples of P2P networks are Kazaa, eDonkey and LimeWire) 4.1c. (Skip if 1.3g≠1) How many free music tracks did you download from peoples' private Internet websites in an average month during 2005? 4.1.d. (Skip if 1.3e≠1) How many MP3 files did you copy from friends or family in an average month during 2005?
(b) Portable hard drive or memory stick
(c) Portable MP3 player
(d) On writable CDs
(e) On writable DVDs Never use Always use 1 2 3 4 5 6 7 8 9 10 Fewer 1 Same 2 [Skip to 4.8]
(b) You did not want to buy the whole album.
(c) You wanted to hear the song before buying it.
(d) The song or album cost too much.
(e) Other. Please specify:____________ Improved 1 Declined 2 Stayed the same 3 Improved 1 Declined 2 Stayed the same 3 Very strong 1 Somewhat strong 2 Moderate 3 Somewhat limited 4 Very limited 5 Very skilled 1 Skilled 2 Somewhat skilled 3 Not very skilled 4 Not at all skilled 5 Demographics
15 to 19 1 20 to 24 2 25 to 34 3 35 to 44 4 45 to 54 5 55 to 64 6 or over 64 7 Less than grade 9 1 Some high school grades 2 Completed high school 3 Some college, technical school (Quebec: CEGEP) 4 Completed college, technical school (Quebec: CEGEP) 5 Some university 6 Graduated university 7 Graduate degree (Masters, Ph.D.) 8 A student 1 Working full-time, which is 35 hours or more per week 2 Working part-time, which is less than 35 hours per week) 3 Not employed at the present time 4 Not able to work 5 A homemaker 6 or retired 7 Other (specify) 77 Full-time 1 Part-time 2 I do not work 3 Under $10 000 $10 000 to less than $20,000 1 $20,000 to less than $40,000 2 $40,000 to less than $60,000 3 $60,000 or more 4
(Read list) Increase 1 Decrease 2 Stay the same 3 Male 1 Female 2 English 1 French 2 Industry Canada Music File Sharing Study 2006
Decima Research Inc.Table of Contents
QRegn. Regions — for quotas
Banner 1 DL/N-DL Region Age Gender Downloader by Region Downloader by Age Age by Gender Total Down
loaderNon-Down
loaderQuebec ROC 25 25+ Men Women DL-
QuebecDL-
ROCN-DL-
QuebecN-DL-
ROCDL-
25DL-
25+N-DL-
25N-DL-
25+ 25
Men25+
Men 25
Women25+
Women
Independent T-Test for Means, Independent Z-Test for Percentages
Upper case letters indicate significance at the 95% level.
"*" Denotes Chi-Square where at least one cell has an expected value of less than 1 or more than 20% of the cells have an expected value of less than 5. (A) (B) (C) (D) (E) (F) (G) (H) (I) (J) (K) (L) (M) (N) (O) (P) (Q) (R) (S) (T) (U) Total 2,100 602 1,498 514 1,586 529 1,571 1,009 1,091 147 454 367 1,131 246 356 283 1,216 280 729 249 842 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Unweighted 'N' 2,100 1,000 1,100 1,006 1,094 1,155 945 1,035 1,065 500 500 506 594 640 360 515 585 577 458 578 487 Atlantic 162 46 116 – 162 41 121 78 84 – 46 – 116 19 27 22 94 22 56 19 65 8% 8% 8% 10% 8% 8% 8% 8% 10% 10% 8% 8% 8% 8% 8% 8% 8% 8% Quebec 514 147 367 514 – 129 385 247 267 147 – 367 – 60 87 69 298 68 179 61 206 24% 24% 24% 100% 24% 24% 24% 24% 100% 100% 24% 24% 24% 24% 24% 24% 24% 24% Ontario 794 227 566 – 794 200 594 381 412 – 227 – 566 93 134 107 459 106 276 94 318 38% 38% 38% 50% 38% 38% 38% 38% 50% 50% 38% 38% 38% 38% 38% 38% 38% 38% Manitoba/ Saskatchewan 173 59 114 – 173 45 128 81 92 – 59 – 114 27 32 18 96 25 56 20 72 8% 10% 8% 11% 9% 8% 8% 8% 13% 10% 11% 9% 7% 8% 9% 8% 8% 9% P Alberta 219 70 149 – 219 67 152 97 121 – 70 – 149 30 40 37 112 33 64 34 88 10% 12% 10% 14% 13% 10% 10% 11% 15% 13% 12% 11% 13% 9% 12% 9% 13% 10% S British Columbia 238 52 186 – 238 47 192 124 114 – 52 – 186 17 35 29 157 25 99 21 93 11% 9% 12% 15% 9% 12% 12% 10% 11% 16% 7% 10% 10% 13% 9% 14% 9% 11% B F K N RT Chi-Square Significance 88%* 100%* 82%* 26%* 100%* 60%* 30%* Q1.3a. Did you buy music CDs during 2005?
Banner 1 DL/N-DL Region Age Gender Downloader by Region Downloader by Age Age by Gender Total Down
loaderNon-Down
loaderQuebec ROC 25 25+ Men Women DL-
QuebecDL-
ROCN-DL-
QuebecN-DL-
ROCDL-
25DL-
25+N-DL-
25N-DL-
25+ 25
Men25+
Men 25
Women25+
Women
Independent T-Test for Means, Independent Z-Test for Percentages
Upper case letters indicate significance at the 95% level. (A) (B) (C) (D) (E) (F) (G) (H) (I) (J) (K) (L) (M) (N) (O) (P) (Q) (R) (S) (T) (U) Total 2,100 602 1,498 514 1,586 529 1,571 1,009 1,091 147 454 367 1,131 246 356 283 1,216 280 729 249 842 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Unweighted 'N' 2,100 1,000 1,100 1,006 1,094 1,155 945 1,035 1,065 500 500 506 594 640 360 515 585 577 458 578 487 Yes 1,618 489 1,129 401 1,216 436 1,182 748 870 120 368 281 848 198 291 238 891 229 518 207 663 77% 81% 75% 78% 77% 82% 75% 74% 80% 82% 81% 77% 75% 80% 82% 84% 73% 82% 71% 83% 79% C G H M M Q Q Q S S S No 479 113 366 113 366 93 386 261 217 27 86 86 280 48 64 44 322 50 211 42 175 23% 19% 24% 22% 23% 17% 25% 26% 20% 18% 19% 23% 25% 20% 18% 16% 26% 18% 29% 17% 21% B F I JK NOP RTU Don't Know/ Refused 4 0 4 0 4 0 4 0 4 0 – – 4 0 – – 4 0 – – 4 *% *% *% *% *% *% *% *% *% *% *% *% *% *% *% Chi-Square Significance 100% 42% 100% 100% 96% 100% 100% Q1.3b. Did you buy music tracks from pay-sites like iTunes or Archambault during 2005?
Banner 1 DL/N-DL Region Age Gender Downloader by Region Downloader by Age Age by Gender Total Down
loaderNon-Down
loaderQuebec ROC 25 25+ Men Women DL-
QuebecDL-
ROCN-DL-
QuebecN-DL-
ROCDL-
25DL-
25+N-DL-
25N-DL-
25+ 25
Men25+
Men 25
Women25+
Women
Independent T-Test for Means, Independent Z-Test for Percentages
Upper case letters indicate significance at the 95% level. (A) (B) (C) (D) (E) (F) (G) (H) (I) (J) (K) (L) (M) (N) (O) (P) (Q) (R) (S) (T) (U) Total 2,100 602 1,498 514 1,586 529 1,571 1,009 1,091 147 454 367 1,131 246 356 283 1,216 280 729 249 842 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Unweighted 'N' 2,100 1,000 1,100 1,006 1,094 1,155 945 1,035 1,065 500 500 506 594 640 360 515 585 577 458 578 487 Yes 275 144 131 62 213 84 191 146 129 39 105 23 108 56 88 28 103 40 106 44 86 13% 24% 9% 12% 13% 16% 12% 14% 12% 26% 23% 6% 10% 23% 25% 10% 8% 14% 15% 18% 10% C G LM LM PQ PQ U No 1,814 456 1,357 452 1,361 444 1,370 858 956 108 348 344 1,013 190 266 254 1,104 239 618 205 751 86% 76% 91% 88% 86% 84% 87% 85% 88% 74% 77% 94% 90% 77% 75% 90% 91% 86% 85% 82% 89% B JKM JK NO NO T Don't Know/ Refused 11 1 10 – 11 1 10 5 6 – 1 – 10 – 1 1 9 – 5 1 5 1% *% 1% 1% *% 1% 1% 1% *% 1% *% *% 1% 1% *% 1% Chi-Square Significance 100% 61% 97% 93% 100% 100% 99% Q1.3d. Did you download free music from P2P file sharing networks, like Kazaa or LimeWire during 2005?
Banner 1 DL/N-DL Region Age Gender Downloader by Region Downloader by Age Age by Gender Total Down
loaderNon-Down
loaderQuebec ROC 25 25+ Men Women DL-
QuebecDL-
ROCN-DL-
QuebecN-DL-
ROCDL-
25DL-
25+N-DL-
25N-DL-
25+ 25
Men25+
Men 25
Women25+
Women
Independent T-Test for Means, Independent Z-Test for Percentages
Upper case letters indicate significance at the 95% level. (A) (B) (C) (D) (E) (F) (G) (H) (I) (J) (K) (L) (M) (N) (O) (P) (Q) (R) (S) (T) (U) Total 2,100 602 1,498 514 1,586 529 1,571 1,009 1,091 147 454 367 1,131 246 356 283 1,216 280 729 249 842 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Unweighted 'N' 2,100 1,000 1,100 1,006 1,094 1,155 945 1,035 1,065 500 500 506 594 640 360 515 585 577 458 578 487 Yes 602 602 – 147 454 246 356 346 255 147 454 – – 246 356 – – 152 195 94 161 29% 100% 29% 29% 47% 23% 34% 23% 100% 100% 100% 100% 54% 27% 38% 19% G I STU U SU No 1,490 – 1,490 367 1,123 282 1,208 655 835 – – 367 1,123 – – 282 1,208 127 528 155 681 71% 99% 71% 71% 53% 77% 65% 77% 100% 99% 100% 99% 45% 72% 62% 81% F H RT R RST Don't Know/ Refused 8 – 8 0 8 1 7 8 0 – – 0 8 – – 1 7 1 7 0 – *% 1% *% 1% *% *% 1% *% *% 1% *% 1% *% 1% *% Chi-Square Significance 100% 4% 100% 100% 100% 100% 100%