Author Topic: Convenience sampling  (Read 659 times)

Offline Tofazzal.ns

  • Sr. Member
  • ****
  • Posts: 314
  • Test
    • View Profile
Convenience sampling
« on: September 19, 2016, 10:47:15 AM »
Convenience sampling is generally known as careless, unsystematic, accidental or opportunistic sampling. The sample is selected according to the convenience of the sample. The researcher selects certain units convenient to him. It requires no pre-planning for the selection of items. Convenience sampling ensures convenience in respect of availability of source list and accessibility of the units. Despite being unscientific, a large number of samples are convenient sampling.
 
A convenience sampling is used in the following situations:
    1. when the universe is not clearly defined
    2. where sampling unit is not clear and
    3. when a complete source list is not available.
Muhammad Tofazzal Hosain
Lecturer, Natural Sciences
Daffodil International University

Offline anam

  • Faculty
  • Full Member
  • *
  • Posts: 128
  • Mathematics transforms everything in Numeric form
    • View Profile
Re: Convenience sampling
« Reply #1 on: November 15, 2016, 06:44:31 AM »
sir please mention the ref. book
Sayedul Anam
Assistant Professor
Department of General Educational Development
Faculty of Science and Information Technology

Offline imranmahmud

  • Full Member
  • ***
  • Posts: 137
  • Test
    • View Profile
Re: Convenience sampling
« Reply #2 on: November 18, 2016, 10:20:39 PM »
For non-probability (if population is unknown), Between convenience and purposive sampling, the purposive sampling would be best. In purposive sampling you can set some rules (sampling frame) to check whether you respondent qualifies for the survey or not which is absent in convenience sampling. These links will explain more

http://dissertation.laerd.com/convenience-sampling.php

http://dissertation.laerd.com/purposive-sampling.php

 
- Imran Mahmud
Assistant Professor,Dept. of SWE
Associate Director (Research)

Offline Bipasha Matin

  • Sr. Member
  • ****
  • Posts: 300
  • Don't judge me, you can't handle half of what I've
    • View Profile
Re: Convenience sampling
« Reply #3 on: November 27, 2016, 09:49:28 AM »
But it is not acceptable in quality research
Sabiha Matin Bipasha

Senior Lecturer
Department of Business Administration
Faculty of Business & Economics
Daffodil International University

Offline imranmahmud

  • Full Member
  • ***
  • Posts: 137
  • Test
    • View Profile
Re: Convenience sampling
« Reply #4 on: November 27, 2016, 02:30:18 PM »
Bipasha Madam, are you referring to Convenience or purposive sampling?
- Imran Mahmud
Assistant Professor,Dept. of SWE
Associate Director (Research)

Offline Bipasha Matin

  • Sr. Member
  • ****
  • Posts: 300
  • Don't judge me, you can't handle half of what I've
    • View Profile
Re: Convenience sampling
« Reply #5 on: December 06, 2016, 12:27:08 PM »
Convenience Sampling.
Sabiha Matin Bipasha

Senior Lecturer
Department of Business Administration
Faculty of Business & Economics
Daffodil International University

Offline Nujhat Anjum

  • Sr. Member
  • ****
  • Posts: 474
  • Test
    • View Profile
Re: Convenience sampling
« Reply #6 on: December 06, 2016, 05:16:57 PM »
Thanks for sharing.

Offline imranmahmud

  • Full Member
  • ***
  • Posts: 137
  • Test
    • View Profile
Re: Convenience sampling
« Reply #7 on: December 07, 2016, 12:46:54 AM »
But still if research think to use convenience sampling, some rules are given below

From Hair et al. 2014 book

PLS-SEM has higher levels of statistical power in situations with complex model structures or smaller sample sizes.
Unfortunately, some researchers believe that sample size considerations do not play a role in the application of PLS-SEM. This idea is
fstered by the often-cited 10 times rule (Barclay, Higgins, & Thompson,
1995),which indicates

the sample size should be equal to the larger of

1. 10 times the largest number of formative indicators used to
measure a single construct, or

2. 10 times the largest number of structural paths directed at a
particular construct in the structural model.


This rule of thumb is equivalent to saying that the minimum
sample size should be 10 times the maximum number of arrowheads
pointing at a latent variable anywhere in the PLS path model. While
the 10 times rule offers a rough guideline for minimum sample size
requirements, PLS-SEM-like any statistical technique-requires
researchers to consider the sample size against the background of the
model and data characteristics (Hair, Ringle, & Sarstedt, 2011).

Specifically, the required sample size should be determined by means
of power analyses based on the part of the model with the largest
number of predictors.
- Imran Mahmud
Assistant Professor,Dept. of SWE
Associate Director (Research)

Offline Bipasha Matin

  • Sr. Member
  • ****
  • Posts: 300
  • Don't judge me, you can't handle half of what I've
    • View Profile
Re: Convenience sampling
« Reply #8 on: December 15, 2016, 10:17:46 AM »
Yes I know that some research book is mentioning this method. But from my personal experience I am asking not to use this method. Last time when we did a research on Foreign Tourists Attitude towards Bangladesh Tourism, and presented the paper in front of some research experts, they were so negative towards our research outcome as they are saying it can't be acceptable findings if we use CONVENIENCE SAMPLING. So later we had to change the whole methodology. :(
Sabiha Matin Bipasha

Senior Lecturer
Department of Business Administration
Faculty of Business & Economics
Daffodil International University

Offline imranmahmud

  • Full Member
  • ***
  • Posts: 137
  • Test
    • View Profile
Re: Convenience sampling
« Reply #9 on: December 15, 2016, 06:04:02 PM »
Hi Madam,

These experts gave us a lot of knowledge so far. Now it is our time to return them something back.

I understand the bias of random sampling from some "old school" professors. I also got rejected from a journal with impact 0.77 for purposive/ convenience sampling but accepted in other journal with impact 1.23(Sage) . But nowadays, there is nothing called old or new school. It is simply "right school" or "wrong school." Logic is simple , if you know your population then chose random, if you don't then go for purposive/ convenience. 

About your paper on foreign tourists you did not have the list of tourists. So random sampling cant be done as you did not know your population. You were right and those experts were wrong. For from your title of the research , I get that you were working with individual level. For organizational level of study, random sampling can be done as you can get the list of organizations somewhere.

My research area is Management Information System . I have a research paper on students online game playing behavior which impact factor is 2.26 (Elsevier) where I used purposive (modified version of convenience sampling) because I did not have the list of students who play Clash of clans games. So my method was acceptable to the reviewers.

I am also giving you a link on tourism paper from the journal called "Tourism Management" from Elsevier, with impact factor 3.1. I can not attach that because of the copyright issue.

Title is "Linking the dots among destination images, place attachment, and
revisit intentions: A study among British and Russian tourists". I have attached their method as image with this message.

http://ac.els-cdn.com/S0261517716302072/1-s2.0-S0261517716302072-main.pdf?_tid=6d151cee-c2b8-11e6-bf2d-00000aacb361&acdnat=1481800984_220b9ec129996b196333e6843f1b0470

If you want, i can email you this paper. Sorry for the long post.

But still you can chose any method you want in your research. They just look for justification.

Thanks
- Imran Mahmud
Assistant Professor,Dept. of SWE
Associate Director (Research)