Тёмный

SPSSisFun: Dealing with missing data (Listwise vs Pairwise) 

SPSSisFun
Подписаться 2,2 тыс.
Просмотров 42 тыс.
50% 1

Опубликовано:

 

20 авг 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 10   
@theempress5058
@theempress5058 5 лет назад
excellent work ,thanks
@piusameh102
@piusameh102 2 года назад
Thanks for this!
@jasoncastledine1012
@jasoncastledine1012 6 лет назад
what if you have multiple independent and dependent variables
@nadak5128
@nadak5128 6 лет назад
Thank you
@jasoncastledine1012
@jasoncastledine1012 6 лет назад
cant fit more than one in the grouping variables ...i have age, gender, school as IVS to 3 different tests on the DV
@user-rq7so2is6n
@user-rq7so2is6n 5 лет назад
May I ask a further question? In linear regression, if we use listwise deletion, would the models by stepwise, forward selection or backward selection be different?
@kingsanalytics2193
@kingsanalytics2193 2 года назад
In listwise, you should not worry about analysis/outcomes, whether linear regression or hierarchical regression or logistic regression. The basic idea is to remove the entire row/respondent data that is affected by one or more missingness. Just an addition - you should be fine if the data is missing at random (MAR), ex, due to mistake omission by respondents. BUT You may face a problem of low statistical power which leads to invalid conclusion , if the respondent(s) made the omission intentionally (maybe due to the fact that you asked sensitive or ambigious question).
@sagwatie
@sagwatie 7 лет назад
does this mean that for a comparison study for example, observed vs estimated, listwise is the best?
@kingsanalytics2193
@kingsanalytics2193 2 года назад
In listwise, you should not worry about analysis/outcomes, whether linear regression or hierarchical regression or logistic regression. The basic idea is to remove the entire row/respondent data that is affected by one or more missingness. Just an addition - you should be fine if the data is missing at random (MAR), ex, due to mistake omission by respondents. BUT You may face a problem of low statistical power which leads to invalid conclusion , if the respondent(s) made the omission intentionally (maybe due to the fact that you asked sensitive or ambigious question).
@bassamalsheakhly1889
@bassamalsheakhly1889 Год назад
Thanks, may i have your e mail? I need help with my data
Далее
Excluding missing data
4:59
Просмотров 63 тыс.
2: Dealing with missing data
9:34
Просмотров 47 тыс.
Replacing Missing Values in SPSS with the Series Mean
12:31
SPSS Missing Values
15:57
Просмотров 58 тыс.
Coding Missing Values in SPSS
11:56
Просмотров 230 тыс.
How to choose an appropriate statistical test
18:36
Просмотров 129 тыс.