Online Exploration: Issues, Issues and an Illustration Procedure.


58 views
Uploaded on:
Description
Various studies, still, run both research center and web tests. ... Behavioral Research on the Internet. Upper Saddle River, NJ: Prentice ...
Transcripts
Slide 1

Online Research: Issues, Problems and an Example Technique John H. Krantz Hanover College

Slide 2

Outline Brief History Issues in Internet Research An Example Method Problems and Solutions

Slide 3

History Email for 2-3 Decades 1995 Krantz, Ballard, & Scher (1997) Within Subjects Reips (1997) Between Subjects First True Web Experiments July 1995 List of Online Psychological Research – supported by APS for a little while http://psych.hanover.edu/research/exponnet.html Almost 200 studies and study locales recorded

Slide 4

Why Do Internet Research Musch & Reips (2000) Sample Size Statistical Power High Speed Ability to achieve members in different nations High outside or natural legitimacy Low cost Ability to recreate a lab explore different avenues regarding more power Special populaces

Slide 5

Issues Is the Sample Representative Yes and No Important to recollect base of correlation Ideal versus Lab Krantz & Dalal (2000) Much more differing that most lab tests Especially in age and training range Race is for the most part constrained as is nationality

Slide 6

Sample Characteristics: Gender Are Internet Samples Male Dominated? Krantz & Dalal (2000) % Female GVU 1 st (1994) 5% Reips (1996) English: 43% German: 18% Krantz, et al. (1997) 44% Pasveer & Ellard (1998) 3 rd Study 71% More Recent Caddell & Utt (2004) 77% Meyerson & Tryon (2003) 45%

Slide 7

Sample Characteristics: Age Are despite everything we testing school sophomores? Krantz, et al. (1997) 43% > 30 Smith & Leigh (1997) 35% > 30 Pasveer & Ellard (1998) 45% > 25 GVU 1 st 36% > 30 Caddell & Utt (2004) 60% > 30 Pattison & Rouse(2003) 16% > 30

Slide 8

Is the College Sophomore be Making a Comeback Plot of action of Psychological Research on the Net Pattison & Rouse (2004) 76% 18-22

Slide 9

Sample Characteristics: Race How different are the specimens? Shockingly when all is said in done examples, differences is still restricted Krantz et al. (1997) 89% White Smith & Leigh (1997) 86% White GVU 10 th (1998b) 87% White O\'Neil, Penrod, & Bornstein (2003) 82% White Meyerson & Tryon (2003) 93% White However, as will be talked about later, web can make it conceivable to get to unique populaces

Slide 10

Sample Characteristics: Nationality Where to do the subjects originate from? To a great extent North American, even US, even in some European studies Krantz, et al. (1997) 86% N. Am. Senior, et al (1999) >80% N. Am This study was directed in England.

Slide 11

Is the Data Any Good Emphatically Yes

Slide 12

Data Quality: Direct Comparisons various concentrates, still, run both research facility and web tests. Krantz, et al. (1997) Regression of web means on lab tests (despite the fact that diverse sorts of tests): lab mean = 1*(Web mean) + 0 r 2 = .99

Slide 13

Compare to Established Data Compare web results to already distributed information sets Myerson & Tryon (2003) Studied Sexual Boredom Scale of Watt & Ewing (1996) Matched example attributes Found same inner consistency Form of organization was not a critical element

Slide 14

Direct Validity Comparison Use same procedures to approve results Pasveer & Ellard (1998) Developed new scale Internal consistency Psychomectric properties

Slide 15

Data Validity: Arguments Reips (2000) Statistical Power Limited Sample Population Limited External Validity Less than Optimal Voluntariness Motivational Confounding Experimenter Bias Nontransparency Limitations of what is doable to inquire about

Slide 16

Interesting Deviations Stern & Faber (1997) Milgram\'s lost letter strategy Milgram, sent on Stern & Faber, came back to sender From doing support to doing simple assignment Often impacts are littler

Slide 17

Getting Subjects: General http://psych.hanover.edu/research/exponnet.html Other Pages: Social Psychology: http://www.socialpsychology.org/expts.htm The Web Experiment List: http://genpsylab-wexlist.unizh.ch/

Slide 18

Getting Subjects: Special Populations Advertising Email Groups Careful, get authorization Can be thought spam Netiquette Discussion Groups Same, consent

Slide 19

Sample Method for Doing Survey Birnbaum, M. H. (2001). Prologue to Behavioral Research on the Internet. Upper Saddle River, NJ: Prentice Hall. Bunches of assets at http://ati.fullerton.edu/NSF supported classes at Fullerton on Social Psychology One more next January, search for the declaration at this website

Slide 20

First Step: Where to put it Setting up a server can be simple Apache web server http://www.apache.org Most well known server Freeware Any PC associated with web is presently a server May have issues of firewalls & dynamic IP addresses

Slide 21

Second Step: SurveyWiz In Birnbaum (2001) Also connected to from http://ati.fullerton.edu/Under Birnbaum\'s page Freely accessible We will manufacture a short overview to outline

Slide 22

Forms SurveyWiz utilizes web structures to gather information What are structures: Elements to permit site page viewers to information This information is then sent back to the server to be prepared SurveyWiz What the page resembles My Take Can alter it with any page editorial manager

Slide 23

Examples I have done for the current year Caddell & Utt Pattison & Reese

Slide 24

Third Step: Collecting the information First way, let Birnbaum do it for you SurveyWiz is set up along these lines <FORM action=http://psych.fullerton.edu/cgi-win/polyform.exe/nonexclusive method=post> FTP://guest:guest99@psych.fullerton.edu You can sort in the location box of your program. Note this is FTP site . At the present time, secret key is guest99, as appeared above, and licenses download as it were.

Slide 25

Keeping the Data Local Second, get this perl script (found at http://psych.fullerton.edu/mbirnbaum/programs/PERL_script2.htm ) The script and guidelines are at this webpage Written by Billy Schmidt Also change code in structure summon At my school the line peruses : <FORM action=http://psych.hanover.edu/cgi-receptacle/survey_wiz.pl method=post>

Slide 26

What is CGI Common Gateway Interface Method for different projects to cooperate with web servers For this situation, this perl program takes information from web structures and stores them in a document Stores information in Excel or SPSS discernable records

Slide 27

Fourth Step: Example Data File Data File Data coding conspire: all variable names start two numbers, see illustration The perl program sorts the variables by numerical request

Slide 28

Another Method of Survey Generation Schmit\'s Survey Assistant He oversees information More adaptability http://or.psychology.dal.ca/~wcs/covered up/home.html

Slide 29

What Do I Need? At NSF ATI site List of programming assets

Slide 30

What Problems are essential Musch & Reips (2000) – 5 point scale No influence over member\'s behavior 3.6 No power over motivation 3.4 Inability of members to ask questions 3.3 Nonrepresentative sample 2.9 Manipulation and fraud 2.4 Ethical problems 1.5

Slide 31

Problems Eliminating Multiple Entries Dropout Security Data Integrity

Slide 32

Multiple Entries People submit too quick, while sitting tight for input People will fake being two individuals Usually have same IP address SurveyWiz and most different techniques, sends the IP location of machine where review is being taken Can wipe out more than one from same IP, e.g. Schmidt (1997)

Slide 33

Security & Data Integrity If on open server, others can get to information and download Others may fake pages to send information (far-fetched) Keep information in non-open catalogs so just analysts have admittance Have CGI check for birthplace of study to ensure it is yours (Schmidt, 1997)

Slide 34

Dropout People will come however not complete Or information is deficient Survey will sends a complete sign to help you track Can utilize: (O\'Neil, Penrod, & Bornstein, 2003) Short sweet Financial motivating forces. In the event that different pages to review, a few pages warm-up so not drop-out amid

Slide 35

Ethical Issues Should not consider Ethical Issues of Web Research Alone Compare to Ethics of doing Traditional Research Not give our solace with what we a chance to have done visually impaired us to those issues

Slide 36

Conclusion Many Benefits Easy Methods Easy Resources Not distinct option for customary strategies another device still should be utilized mindfully

Recommended
View more...