Dissertation Statistics Helper

Stockholm Syndrome in Dissertation Students?

Stockholm Syndrome is a condition in which those held hostage by their captors develop an alliance with those captors. The captors in these circumstances hold all the cards. They control behavior through their power to punish, but they are also the benevolent distributors of positive reinforcement—food, water, blankets, etc. Stockholm Syndrome was first noted in 1973 when four individuals were taken hostage in a bank robbery in Stockholm, Sweden. After being released, the hostages would not testify against their captors and even defended the robbers' actions.

But Stockholm Syndrome isn’t just a phenomenon seen in those who were taken hostage by bank robbers or terrorists. It is not uncommon to see it in doctoral students who are in the throes of the dissertation process. I have worked on the doctoral dissertations of about a dozen students who sought their doctoral degrees from one of the large American universities that specialize in online doctorates. That school’s process is soul crushing, consisting of multiple reviews, each performed by individuals who have absolute authority to stop the dissertation dead in its tracks if they aren’t pleased or feel that they haven't been sufficiently kowtowed to. Absolute power corrupts absolutely, and that principle is manifested clearly in the behavior of these reviewers. Many times these people are as thoroughly confident in their ability as they are, in reality, incompetent. But they can demand anything, no explanation or justification needed, and the student must deliver. Or else the dissertation process stops. And when the next reviewer in the endless series of reviews demands just the opposite (yes, I’ve seen that), the student must deliver. Or else. And this continues until the university has squeezed the student dry—financially and emotionally—and moves on to the next victim.

The remarkable thing is that I’ve seen students working through the dissertation process at this institution (and, to be fair, at some other schools as well) who sided with their captors, attempting to explain the wisdom of the process to me, and justifying the abuse they received by pointing out that it was balanced by the most benevolent encouragement from the same reviewers who created the need for that encouragement in the first place! Good cop. Bad cop. All the same cop.

When I have suggested to some of these students that they’re the victims of academic bullying and manipulation, a few reacted defensively or even angrily. One such client recently terminated our consulting relationship me when I suggested that some push back was called for against a committee member who was requiring multiple reviews when one would do, demanding that statistical results be moved to chapter 5 from chapter 4 where they belonged, insisting that one statistic be substituted for an equivalent statistic (no justification offered), and making multiple other whimsical demands--whatever ticked his fancy at the moment it seemed. Stockholm Syndrome? I think so.

Wouldn't this make an interesting dissertation?! Good luck finding a committee, though!

G*Power does not provide any direct application for estimating sample size requirements for partial or semipartial (“part”) correlation analysis. Various workarounds are suggested online, ranging from the impossibly complicated to simply ignoring the covariates and using the same app within G*Power that is used for the Pearson correlation.

The procedure I favor is to use the G*Power app for estimating sample size (in a priori power analysis) or power (in post hoc power analysis) for tests of the significance of the individual predictors in multiple linear regression. This approach is based on the fact that the tests of significance of the individual predictors in SPSS multiple regression analysis are also the tests of the significance of the related partial and semipartial correlations.

Here is the G*Power procedure:

Tests > Correlation and regression > Linear multiple regression: Fixed model, single regression coefficient

Type of power analysis: a priori

Tail(s): chose one-tail if you have predicted the sign of the partial or semipartial correlation; chose two-tail if you’ve not made a prediction or are interested in a correlation in either direction

Effect size f-squared: Use .02 if you want to detect a weak population correlation, .15 for a medium population correlation, and .35 for a strong population correlation

Alpha err prob: your chosen level of significance, typically .05

Power (1 – beta err prob): your chosen level of statistical power, typically .80 which will give you a Type II error probability of .20

Number of predictors: set equal to the total number of variables in your partial or semipartial correlation analysis (X and Y and all the covariates) minus one.

If you’d like to see some additional tips and techniques, I hope you’ll visit my website: www.DissertationStatsHelper.com

If you need statistical consulting work, email me at [email protected]. (Please note that I don’t check FaceBook for messages.)

Dissertation Statistics Helper A PhD author of two statistics textbooks will guide you to a successful dissertation. Contact me today for a free initial consultation!

Should Multi-Item Inventories Contain Reverse-Worded Items?

One common practice used to control for one form of response bias when collecting rating scale data is to reverse-word some items. Survey respondents who are presented with a series of items that are all positively worded (e.g., "I think that statistics is fun." "I have a good time working with statistics." "I enjoy reading about statistics.") can lead some respondents to start checking all high (or all low) ratings without really thinking about what they're doing. By reverse-wording some of the items (e.g., "I do not think that statistics is fun." "I do not have a good time working with statistics." "I do not enjoy reading about statistics.") the idea is that respondents are forced to read the items more closely which should lead them to make more informed ratings. Of course reverse-worded items must be reverse-scored before total scores are calculated.

While this seems like a good idea in theory, in practice I've observed something else when inventories are constructed of a mixture of positively-worded and negatively-worded items. Factor analyses of these inventories frequently finds two factors--one on which positively-worded items load strongly, and one on which negatively-worded items load strongly. That means that the internal consistency of the inventory is challenged and that means the inventory totals (or averages) are invalid.

Perhaps a more desirable approach to breaking response bias is to mix together the items from two (or more) inventories measuring very different constructs. Word items in both inventories in a positive direction, but force respondents to slow down and read the items by ensuring that any two consecutive items deal with different issues.

If you are working on a dissertation, thesis, or DNP capstone research project and need a methodologist, I hope you'll check my website: www.DissertationStatsHelper.com

Dissertation Statistics Helper Statistical consulting for students working on doctoral dissertations and masters theses.

Every so often changes are made in SPSS syntax that can catch one by surprise. For instance, a couple versions back, one could write a couple simple lines of syntax to specify that analyses beyond that point would run on only a select group of cases.

For instance, if you wanted to run an analysis on only cases whose age was less than 30, you could type these two lines in a syntax file and run them:

SELECT IF (AGE LT 30).
EXECUTE.

Those two lines of syntax would simply tag unselected cases (i.e., cases who were 30 and older) and not include them in the analysis, but the cases still remained in the data file.

At some point in the recent past, however, the rules changed! Running those same two lines now doesn't just exclude the cases from the analysis, but DELETES THEM FROM THE DATA FILE!!

The syntax required to select a subset of cases for analysis but still leave the other cases in the data file is a lot more complex. For the example begun above, where you wanted to select cases for subsequent analysis who were less than 30 years old:

USE ALL.
COMPUTE filter_$=(age lt 30).
VARIABLE LABELS filter_$ 'age lt 30 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

Yikes! I recommend that you use the menu system to generate the necessary syntax and then paste it into your syntax file in the appropriate location:

Data > Select Cases > Check "If condition is satisfied" > Click the "If..." button > type in "age lt 30" (without the quotes) > Click the "Continue" button > Click the "OK" button.

If you are looking for assistance with research design or statistical analysis on your dissertation or thesis research, I'd invite you to visit my website, www.DissertationStatsHelper.com Perhaps I can help.

www.dissertationstatshelper.com