BRENAN DALE D. SENDICO
Master of Arts in Education – Administration & Supervision
CENTRAL MINDANAO UNIVERSITY
Research Methods in Education (Educ241)
INTERNET – BASED SURVEYS
1. EMAIL SURVEY
• An email survey is one that sends the survey instrument (e.g.
questionnaire) to a respondent via email and most often samples
respondents via email.
2. Web-Based Surveys
• Web surveys allow respondents to complete questionnaires that are
delivered to them and administered over the World Wide Web.
• Sends direct potentials participants to a link or web site at which the
survey questionnaire is located in HTML form.
INTERNET – BASED SURVEYS
•The most widely used data collection
instrument for Internet surveys is the
QUESTIONNAIRE.
INTERNET – BASED SURVEYS
The following are advantages in using an Internet questionnaire in
comparison to a paper questionnaire (e.g Watt 1997; Dillman et al.
1999; Dillman and Bowket 2000; Roztocki and Labri 2002)
1. It reduces cost
2. It reduces the time to take to distribute ,gather and
process data.
3. It enables a wider and much larger population to be
accessed.
4. It enables researchers to reach difficult population
under the cover of anonymity and non-traceability
5. It may have novelty value
6. Respondents can complete the questionnaire from
home.
7. Respondents can complete it at a time to suit
themselves.
INTERNET – BASED SURVEYS
8. Respondents can complete the survey over time.
9. Reduction of researcher effects.
10. Responses in web-based surveys show fewer missing
entries than paper-based surveys.
11. Human error is reduced in entering and processing
online data.
12. Additional features may make the survey attractive.
INTERNET – BASED SURVEYS
13. Greater generalizability may be obtained as Internet
users come from a wide and diverse population.
14. Because of volunteer participation, greater
authentication of responses may be obtained
INTERNET – BASED SURVEYS
INTERNET – BASED SURVEYS
The importance of the visual aspect of questionnaire is
heightened in Internet surveys ( Smyth et al. 2004)
Smyth et al. (2004) reported that respondents use ‘pre-
attentive processing’ when approaching Internet surveys
which rests on Gestalt psychology that abides by the
principles of:
1.Proximity 2. Similarity 3. Prägnanz
KEY ISSUES IN INTERNET–BASED SURVEYS
Gestalt psychology
1. Proximity - grouping those items that are physically
close to each other
2. Similarity - grouping together those items that
appear alike
3. Prägnanz - figures or items with simplicity, regularly
and symmetry are more easily perceived and
remembered
INTERNET–BASED SURVEYS
PROBLEMS and SOLUTIONS
Sampling
•Some subsample
groups may be under-
represented in the
respondents
Solutions
•Adjust the results by
weighing the sample
responses.
PROBLEMS and SOLUTIONS
Ethics
•Respondents may
wish to keep their
identity from the
researcher , and an
email address
identifies the
respondent.
Solutions
•Direct respondents to
a web site rather than
to using email
correspondence.
PROBLEMS and SOLUTIONS
Technical Hardware &
Software
•questionnaire may
vary from one
machine to another
and can lead to
dropout.
Solutions
•Test the survey on
different computer
systems/browser to
ensure consistency.
Avoid survey that
require real time
completion.
PROBLEMS and SOLUTIONS
• Slow network connection or
limited bandwidth can slow
down loading.
• Respondents may not have the
same software or the same
version of the software as the
sender.
• Keep the use of graphics to
a minimum. Advise on the
possible time it takes to
load.
• Avoid the use of graphics
and more advanced
software programs.
PROBLEMS and SOLUTIONS
• The greater the use of graphics
and plug- ins the longer it
takes to download.
• There may be slow loading
times due to Internet
congestion
• Keep software
requirements as low-tech
as possible
• Avoid sophisticated
graphics and ‘’fancy’’
presentation.
PROBLEMS and SOLUTIONS
• Email questionnaires may
distort the layout of the
questionnaire
• Avoid sending a
questionnaire directly using
email rather post it on a
web site
PROBLEMS and SOLUTIONS
• Respondents may be
unfamiliar or inexperienced
with the internet and the
media
• There may be more than one
respondent to a single
questionnaire
• Keep the questionnaire
simple and easy to
complete.
• Include questions to
crosscheck the consistency
of replies to similar items.
RESPONDENTS SOLUTIONS
PROBLEMS and SOLUTIONS
• Respondents may be
bombarded with too much
information in an introductory
messages
• Respondents may not
understand instruction
• Separate the questionnaire
from the advertisement for
or introduction to the
questionnaire.
• Provide helpline
LAYOUT & PRESENTATION SOLUTIONS
PROBLEMS and SOLUTIONS
• Respondents may alter the
instrument itself.
• Respondents may be forced to
answer every question even
when they consider some
response categories
inappropriate.
• Include technological
safeguard to prevent
alteration and have
procedures to identify
altered instruments.
• Include options such as
‘don’t know’ and ‘do not
wish to answer’
RELIABILITY SOLUTIONS
PROBLEMS and SOLUTIONS
• Respondents may lose interest
after a while and abandon the
survey, thereby losing all the
survey data.
• Respondents may not know
how long the questionnaire is,
and may lose interest
• Have a device that requires
respondents to send their
replies screen by screen.
• Include a device for
indicating how far through
the questionnaire the
respondent has reached
DROPOUT SOLUTIONS
PROBLEMS and SOLUTIONS
• It is easier for someone to quit
an internet – based survey
than a paper- based survey.
• Non participation may be high
• Error messages cause
frustration and may cause
respondents to abandon the
questionnaire
• Increase incentives to
participate
DROPOUT SOLUTIONS
PROBLEMS and SOLUTIONS
Internet-based surveys are subject to
the same ethical rules as
paper-based surveys
Sampling bias is a major concern for Internet based
surveys. Internet-mediated research is immediately
subject to serious problems concerning sampling
representatives and validity of data( Coober 1997;
Roztoki and Lahri 2002)
SAMPLING in Internet-based Surveys
SAMPLING in Internet-based Surveys
Watt ( 1997) suggests the three types of Internet
sample:
1. Unrestricted sample -anyone can complete the
questionnaire, but it may have limited
representatives.
2. Screened sample - quotas are placed on the
subsample categories and types.
SAMPLING in Internet-based Surveys
3. Recruited sample - respondents complete a
preliminary classification questionnaire and then ,
based on the data provided in them, are recruited or
not.
SAMPLING in Internet-based Surveys
High Hurdle Technique ( Reips 2002) to increase
response rate in Internet-based survey
• patience : loading time and long texts
• duration
• Seriousness
• Personalization
• impression of control
SAMPLING in Internet-based Surveys
• Privacy
• Preconditions
• Technical pretest
• Rewards
INTERNET – BASED EXPERIMENTING
Hewson et al. (2003) classify the four principal types of
Internet-based experiments:
1. Static printed materials - Sends formulated material
to respondents (e.g graphically presented material)
2. Use of non-printed material (e.g video clips or
sound)
INTERNET – BASED EXPERIMENTING
3. Reaction- time experiments - requires very precise
timing.
4. Internet interviewing - facilitated by chat rooms.
INTERNET – BASED EXPERIMENTING
DOS in INTERNET – BASED EXPERIMENTING
• use dropout as a dependent variable
• use dropout to detect motivational confounding
• place questions for personal information at the
beginning of the Internet study
• use technique that help ensure quality of data
collection over the Internet.
• use Internet- based tools and services to develop and
announce your study.
DON’TS in INTERNET – BASED
EXPERIMENTING
• Do not allow external access to unprotected
directories
• do not allow public display of confidential participant
data through URLs
• Do not accidentally reveal the experiment’s structure
• Do not ignore the technical variance inherent in the
Internet
DON’TS in INTERNET – BASED EXPERIMENTING
• do not bias the results through improper use of form
elements, such as measurement errors where
omitting particular categories.
INTERNET – BASED INTERVIEWS
INTERNET – BASED INTERVIEWS
• The opportunity that Internet interviews present for
interviewing respondents is immense.
• It enable researchers to contact hard-to-reach groups
and individuals.
• Chat rooms provide opportunity for split screens and
shared screens, thereby displaying the ongoing
dialogue between participants.
INTERNET – BASED INTERVIEWS
• Internet interviewing can also go offline
• It requires both parties to agree a time to log on to
the computer in order to conduct the interview
Internet - Based Research

Internet - Based Research

  • 1.
    BRENAN DALE D.SENDICO Master of Arts in Education – Administration & Supervision CENTRAL MINDANAO UNIVERSITY Research Methods in Education (Educ241)
  • 3.
    INTERNET – BASEDSURVEYS 1. EMAIL SURVEY • An email survey is one that sends the survey instrument (e.g. questionnaire) to a respondent via email and most often samples respondents via email. 2. Web-Based Surveys • Web surveys allow respondents to complete questionnaires that are delivered to them and administered over the World Wide Web. • Sends direct potentials participants to a link or web site at which the survey questionnaire is located in HTML form.
  • 4.
    INTERNET – BASEDSURVEYS •The most widely used data collection instrument for Internet surveys is the QUESTIONNAIRE.
  • 5.
    INTERNET – BASEDSURVEYS The following are advantages in using an Internet questionnaire in comparison to a paper questionnaire (e.g Watt 1997; Dillman et al. 1999; Dillman and Bowket 2000; Roztocki and Labri 2002) 1. It reduces cost 2. It reduces the time to take to distribute ,gather and process data. 3. It enables a wider and much larger population to be accessed.
  • 6.
    4. It enablesresearchers to reach difficult population under the cover of anonymity and non-traceability 5. It may have novelty value 6. Respondents can complete the questionnaire from home. 7. Respondents can complete it at a time to suit themselves. INTERNET – BASED SURVEYS
  • 7.
    8. Respondents cancomplete the survey over time. 9. Reduction of researcher effects. 10. Responses in web-based surveys show fewer missing entries than paper-based surveys. 11. Human error is reduced in entering and processing online data. 12. Additional features may make the survey attractive. INTERNET – BASED SURVEYS
  • 8.
    13. Greater generalizabilitymay be obtained as Internet users come from a wide and diverse population. 14. Because of volunteer participation, greater authentication of responses may be obtained INTERNET – BASED SURVEYS
  • 9.
    INTERNET – BASEDSURVEYS The importance of the visual aspect of questionnaire is heightened in Internet surveys ( Smyth et al. 2004) Smyth et al. (2004) reported that respondents use ‘pre- attentive processing’ when approaching Internet surveys which rests on Gestalt psychology that abides by the principles of: 1.Proximity 2. Similarity 3. Prägnanz
  • 10.
    KEY ISSUES ININTERNET–BASED SURVEYS Gestalt psychology 1. Proximity - grouping those items that are physically close to each other 2. Similarity - grouping together those items that appear alike 3. Prägnanz - figures or items with simplicity, regularly and symmetry are more easily perceived and remembered
  • 11.
  • 12.
    PROBLEMS and SOLUTIONS Sampling •Somesubsample groups may be under- represented in the respondents Solutions •Adjust the results by weighing the sample responses.
  • 13.
    PROBLEMS and SOLUTIONS Ethics •Respondentsmay wish to keep their identity from the researcher , and an email address identifies the respondent. Solutions •Direct respondents to a web site rather than to using email correspondence.
  • 14.
    PROBLEMS and SOLUTIONS TechnicalHardware & Software •questionnaire may vary from one machine to another and can lead to dropout. Solutions •Test the survey on different computer systems/browser to ensure consistency. Avoid survey that require real time completion.
  • 15.
    PROBLEMS and SOLUTIONS •Slow network connection or limited bandwidth can slow down loading. • Respondents may not have the same software or the same version of the software as the sender. • Keep the use of graphics to a minimum. Advise on the possible time it takes to load. • Avoid the use of graphics and more advanced software programs.
  • 16.
    PROBLEMS and SOLUTIONS •The greater the use of graphics and plug- ins the longer it takes to download. • There may be slow loading times due to Internet congestion • Keep software requirements as low-tech as possible • Avoid sophisticated graphics and ‘’fancy’’ presentation.
  • 17.
    PROBLEMS and SOLUTIONS •Email questionnaires may distort the layout of the questionnaire • Avoid sending a questionnaire directly using email rather post it on a web site
  • 18.
    PROBLEMS and SOLUTIONS •Respondents may be unfamiliar or inexperienced with the internet and the media • There may be more than one respondent to a single questionnaire • Keep the questionnaire simple and easy to complete. • Include questions to crosscheck the consistency of replies to similar items. RESPONDENTS SOLUTIONS
  • 19.
    PROBLEMS and SOLUTIONS •Respondents may be bombarded with too much information in an introductory messages • Respondents may not understand instruction • Separate the questionnaire from the advertisement for or introduction to the questionnaire. • Provide helpline LAYOUT & PRESENTATION SOLUTIONS
  • 20.
    PROBLEMS and SOLUTIONS •Respondents may alter the instrument itself. • Respondents may be forced to answer every question even when they consider some response categories inappropriate. • Include technological safeguard to prevent alteration and have procedures to identify altered instruments. • Include options such as ‘don’t know’ and ‘do not wish to answer’ RELIABILITY SOLUTIONS
  • 21.
    PROBLEMS and SOLUTIONS •Respondents may lose interest after a while and abandon the survey, thereby losing all the survey data. • Respondents may not know how long the questionnaire is, and may lose interest • Have a device that requires respondents to send their replies screen by screen. • Include a device for indicating how far through the questionnaire the respondent has reached DROPOUT SOLUTIONS
  • 22.
    PROBLEMS and SOLUTIONS •It is easier for someone to quit an internet – based survey than a paper- based survey. • Non participation may be high • Error messages cause frustration and may cause respondents to abandon the questionnaire • Increase incentives to participate DROPOUT SOLUTIONS
  • 23.
    PROBLEMS and SOLUTIONS Internet-basedsurveys are subject to the same ethical rules as paper-based surveys
  • 24.
    Sampling bias isa major concern for Internet based surveys. Internet-mediated research is immediately subject to serious problems concerning sampling representatives and validity of data( Coober 1997; Roztoki and Lahri 2002) SAMPLING in Internet-based Surveys
  • 25.
    SAMPLING in Internet-basedSurveys Watt ( 1997) suggests the three types of Internet sample: 1. Unrestricted sample -anyone can complete the questionnaire, but it may have limited representatives. 2. Screened sample - quotas are placed on the subsample categories and types.
  • 26.
    SAMPLING in Internet-basedSurveys 3. Recruited sample - respondents complete a preliminary classification questionnaire and then , based on the data provided in them, are recruited or not.
  • 27.
    SAMPLING in Internet-basedSurveys High Hurdle Technique ( Reips 2002) to increase response rate in Internet-based survey • patience : loading time and long texts • duration • Seriousness • Personalization • impression of control
  • 28.
    SAMPLING in Internet-basedSurveys • Privacy • Preconditions • Technical pretest • Rewards
  • 29.
    INTERNET – BASEDEXPERIMENTING Hewson et al. (2003) classify the four principal types of Internet-based experiments: 1. Static printed materials - Sends formulated material to respondents (e.g graphically presented material) 2. Use of non-printed material (e.g video clips or sound)
  • 30.
    INTERNET – BASEDEXPERIMENTING 3. Reaction- time experiments - requires very precise timing. 4. Internet interviewing - facilitated by chat rooms.
  • 31.
    INTERNET – BASEDEXPERIMENTING
  • 32.
    DOS in INTERNET– BASED EXPERIMENTING • use dropout as a dependent variable • use dropout to detect motivational confounding • place questions for personal information at the beginning of the Internet study • use technique that help ensure quality of data collection over the Internet. • use Internet- based tools and services to develop and announce your study.
  • 33.
    DON’TS in INTERNET– BASED EXPERIMENTING • Do not allow external access to unprotected directories • do not allow public display of confidential participant data through URLs • Do not accidentally reveal the experiment’s structure • Do not ignore the technical variance inherent in the Internet
  • 34.
    DON’TS in INTERNET– BASED EXPERIMENTING • do not bias the results through improper use of form elements, such as measurement errors where omitting particular categories.
  • 35.
  • 36.
    INTERNET – BASEDINTERVIEWS • The opportunity that Internet interviews present for interviewing respondents is immense. • It enable researchers to contact hard-to-reach groups and individuals. • Chat rooms provide opportunity for split screens and shared screens, thereby displaying the ongoing dialogue between participants.
  • 37.
    INTERNET – BASEDINTERVIEWS • Internet interviewing can also go offline • It requires both parties to agree a time to log on to the computer in order to conduct the interview