Jawapan Final 29 Nov 2010 - Soalan 3

SOALAN 3

a.       Terangkan persampelan bolasalju dan persampelan kuota

Snowball

•         Used when elements in population have specific characteristics or knowledge, but are very difficult to locate and contact.

•         Initial sample group can be selected by probability or non-probability methods, but new subjects are selected based on information provided by initial subjects.

Eg, used to locate members of different stakeholder groups regarding their opinions of a new public works project

Quota

§  Quotas for numbers or proportion of people to be sampled, established.

Examples:

1)      Survey for research on dual career families: 50% working men and 50% working women surveyed.

2)      Women in management survey: 70% women surveyed and 30% men surveyed.

b.      Apakah yang dimaksudkan dengan “Ketepan” dan “Keyakinan” di dalam mereka bentuk sampel?

§     Precision

ú    refers to how close the sample estimate  eg  X is to the true population characteristic    ( m) depends on the  variablity in the sampling distribution of the mean, ie the standard error  ( S_X )

ú    indicates the confidence interval within which the population mean can be estimated (m= X + KS _X )

§    Confidence

ú   reflects the level of certainty that the sample estimates will actually hold true for the population

ú   bias is absent from the data

ú   accuracy is reflected by the confidence level ( K )

c.       Senaraikan EMPAT (4) unit-unit analisis di dalam rekabentuk penyelidikan.

Senaraikan 4 daripada unit-unit analisa berikut:

1.             Individuals/individu

2.             Dyads/dyad

3.             Groups/kumpulan

4.             Organisations/organisasi

5.             Cultures/budaya

d.      Kaedah pengumpulan data secara temuduga melalui telefon sering digunakan. Kenalpasti TIGA (3) kelebihan dan TIGA (3) kekurangan yang terdapat pada kaedah ini.

 e.       Bincang dengan ringkas EMPAT (4) panduan untuk pembentukan soalan dalam sesuatu

           soalselidik.

a)      Avoid complexity: Simpler language is better.

b)      Avoid leading and loaded questions.

c)      Avoid ambiguity: Be as specific as possible.

d)     Avoid double-barreled items.

e)      Avoid making assumptions.

f)       Avoid burdensome questions that may tax the respondent’s memory.

g)      Make certain questions generate variance.