+ Reply to Thread
Results 1 to 7 of 7

Thread: Very Small Sample Size -What multivariate analysis should be used

  1. #1
    Points: 136, Level: 2
    Level completed: 72%, Points required for next Level: 14

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Smile Very Small Sample Size -What multivariate analysis should be used




    Dear all,

    Good day!

    I'm having a hard time researching for possible multivariate analysis to be used in my study. I have less than 10 sample for the study which is already the whole population. I tried to expand by including other cities in my study, yet they immaterially affected the sample size. The problem is I want to do a correlational analysis by finding the relationship of my 15 independent variables (categorized into 3) to my 1 dependent variable. Could I possibly use multiple regression or other type of regression?

  2. #2
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Very Small Sample Size -What multivariate analysis should be used

    Could you tell us a bit more about your study?
    Topic, research questions? If you really have
    n < 10 then you probably cannot do much more
    than to perform 15 (or 3?) bivariate analyses
    of the relationship between single IVs and the
    AV.

    Kind regards

    K.

  3. #3
    Points: 136, Level: 2
    Level completed: 72%, Points required for next Level: 14

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Very Small Sample Size -What multivariate analysis should be used

    My study centers on the "factors affecting the profitability of small-scale funeral homes in our city". The major constraint is that only few small scale funeral homes exist in our city. I can't see expansion of scope as an alternative way to increase the sample size since it will make my study costly while the effect on the sample size is immaterial (that is to say only 1 or 2 respondent-firm would likely add to the sample size).

    I have one dependent variable which is Return on Equity and 15 independent variables clustered as follow:

    Factors Related to Organizational Resources
    o Current Ratio
    o Ratio of Fixed Assets to Total Assets
    o Debt-to-Asset Ratio
    o Level of Expenses
    o Average Price of Funeral Service
    Factors Related to Operations
    o Efficiency
    o Effectiveness
    o Productivity
    o Number of tie-up Agreements with Life insurance Plans and others
    Factors Related to Policy
    o Inventory Turnover in days
    o Average Collection Period
    o Average Payment Period

    My goal is to know which of the following IVs significantly affect my DV.

    Thanks for entertaining my question. God Bless you all!

  4. #4
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Very Small Sample Size -What multivariate analysis should be used

    The prior post is probably your only option. Conduct descriptive statistics and some bivariate statistics. With the sample size you reported, the data could not support multivariate procedures.

  5. The Following User Says Thank You to hlsmith For This Useful Post:

    ilyes (02-20-2017)

  6. #5
    Points: 136, Level: 2
    Level completed: 72%, Points required for next Level: 14

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Very Small Sample Size -What multivariate analysis should be used

    Hi Sir!

    Thanks for your advice!

    Can I independently regress each of the IVs to my DV to get whether they are significant or not, as well as to know the extent (coefficient) each IV impact my DV? Say for example, regress Current Ratio (IV) with ROE (DV), get the estimated equation, thenafter, another regression for other IVs.

    Thank you very much! Sorry I'm really terrible when it comes to statistics but I can handle analysis very well.

  7. #6
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Very Small Sample Size -What multivariate analysis should be used

    For a sample as small as yours, I would rather recommend using the
    Spearman rank correlation coefficient. You can test that coefficient
    for significance.

    The problem is that you want to perform 15 tests. In that case
    ususally a correction for multiple testing is made, e.g. set the
    level for a "significant" effect from p < 5% to p < 1%.

    Kind regards

    K.

  8. The Following User Says Thank You to Karabiner For This Useful Post:

    ilyes (02-20-2017)

  9. #7
    Points: 160, Level: 3
    Level completed: 20%, Points required for next Level: 40

    Posts
    5
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Re: Very Small Sample Size -What multivariate analysis should be used


    Quote Originally Posted by hlsmith View Post
    The prior post is probably your only option. Conduct descriptive statistics and some bivariate statistics. With the sample size you reported, the data could not support multivariate procedures.
    Hello i have the same problem with my study, could i use multivariate analysis in my study . I have 5 sample for the study which is already the whole population? if not what i should use as an alternative way in order to examine the relation betwen the IV and DV?

    Ps: my study is about the impact of corporate governance on firm performance
    i have 6 IV and 1 DV

    Thanks for ur help in advance

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats