Why Dr. Sokol’s Study May Be Fundamentally Flawed in Regard to the Yeshiva Community


By Rabbi Yair Hoffman for 5tjt.com

In the November 18th, 2022 edition of the Journal for the Scientific Study of Religion (https://onlinelibrary.wiley.com/doi/abs/10.1111/jssr.12812), a study was published by Dr. Yosef Sokol, Naomi Rosenbach,Chayim Rosensweig,Chynna Levin,Shifra Hubner, and Isaac Schechter entitled, “Examining Average Age at First Marriage within Orthodox Judaism: A Large Community-Based Study.” The research has been lauded as debunking “the supposed surplus of eligible women in relation to eligible men.”

Firstly, before we get into the questions on Dr.  Sokol’s data collection and methodologies, I just want to state that I believe Dr. Sokol is a remarkable individual who is a true Ben Torah with exemplary midos.  He is community minded, means well, and seriously wants to help by using his formidable skills and training to help the observant Torah community.  In general, I found the study to be impressive in regard to the general Modern Orthodox Jewish community.  I believe, however, that his study and his reported findings may be fundamentally inaccurate – specifically regarding the Yeshiva and Bais Yaakov communities.

In the first articles that were released about the study, it was reported that 5000 out of a total of 9000 respondents were self-identified family members of the Yeshiva community.  In this author’s opinion, many readers took this study to mean that Dr. Sokol feels that there is no Shidduch crisis.  This is a problem because it undermines the efforts of some very good people who have dedicated themselves to resolving the underlying situation.

Both data collection and the science of conducting studies are not my areas of expertise, but this does not mean that we cannot explore and question.  I did interview Dr. Sokol and presented my questions on his methodology and data collection.


Firstly, the categories that were used in the self-identification section the study might be misleading and may perhaps be given to very serious error.   The categories he used in his study were:

  • 1 Yeshiva Orthodox
  • 2 Modern Orthodox
  • 3 Hasidic
  • 4 Chabad
  • 5 non-Orthodox

I humbly suggest that if one were to Google both the term, “Yeshivish” (549,000 hits) and the term “Yeshiva Orthodox” (1230 hits) that person would conclude that “Yeshiva Orthodox” can be a very misleading term. And most, if not all of those “Yeshiva Orthodox” are not reflective of the actual term.  In other words, the term Yeshiva Orthodox does not exist.  Thus, those who responded with a self-identification in the first category may have felt that choice #1 includes both the category of “Yeshivesh” and the category of “Orthodox” as opposed to the non-existent “Yeshiva Orthodox” as a category.

In other words, it may have been read as Yeshiva or orthodox, two different categories that stand to the right of Modern Orthodox.


The second observation is that if I were to have conducted this study, I would preferred to have analyzed the specific data as it comes from each of the Yeshiva community schools or Bais Yaakovs.  How many of the class of 2010 Bais Yaakov X are married?  Why go through the indirect method of internet surveys discussing other members of the family?  When I posed this question, Dr. Sokol responded, “That was the very first thing that I tried to do.  The Bais Yaakovs, however, did not agree to give over their class lists and therefore developed alternative methods of finding the data.”  It is clear that Dr. Sokol does admit that the data collection was not ideal.  The study itself states (p. 714): ..this study utilized a nonrandomized method of sampling: a crowdsourcing/convenience sampling technique.  The reason that was given was that “there is no standard and available method of random sampling in the Orthodox Jewish community in North America, nor a relevant database from which to conduct random sampling.”


The third observation is that the study states (p. 715) that the survey was distributed and publicized online through paid email advertisements, social media forums, texting platforms, word of mouth, web-based community news forums, and emails from Orthodox Jewish clergy leadership organizations (e.g., Rabbinical Council of America) to their members and affiliated Orthodox Jewish communities.

It is this author’s view, however, that a not insignificant percentage of the Yeshivish Torah community do not visit internet websites or WhatsApp ( even if it is just one member of their family), and the data may be unreflective of the actual situation on the ground.  When we combine this with the first observation, the results can be skewed even more.


The fourth observation is that those who have studied the issue of the Female Shidduch Crisis in the Yeshiva/Bais Yaakov communities are well aware that it is not merely anecdotal.  It is based upon empirical analyses of data that come from schools.  Also, the fact that there is a larger pool of girls because of earlier dating norms is a reality.  These norms do not exist in the Chassidish communities, nor in more modern communities nor in Eretz Yisroel. In response to this point Dr. Sokol explained that those who have studied this issue have not studied whether the same norm exists in the Yeshivos for men. Dr. Sokol claims that the data from the schools does not extend to age 40.

Dr. Sokol disagreed with each of the four observations.  He claimed that the data from the schools up to 30 year-olds is almost identical to what he had found.  His data states that is that when the Yeshiva Orthodox are in their 30’s – another 6 percent of girls do get married, bringing the total Yeshiva Orthodox by age 40 to 96%.


It could be that the aforementioned observations are incorrect, as it is not my area of expertise. However, I would like to note that shadchanim, mothers of boys and girls, and the numerous older girls that I have encountered – point to the reality of this crisis.  The response to this is that all of it is anecdotal.  So what is the plus side?  It could be that Dr. Sokol’s study will give further hope to those who have given up, and encourage them to strengthen their efforts in both hishtadlus and Tefillos to find their bashert.  At the same time, we must strengthen our efforts in trying somehow to minimize the age gap which is making a difficult situation significantly worse.  And to this, both Dr. Sokol and those who question the methodology both agree.

Dr. Sokol responded,  “There is definitely a communal problem of women not getting married when they want to. One take-home message here is that if almost all young men are getting married quickly and a large percentage of young women never get married that presents certain necessary solutions. However, the study showed that both men and women are having trouble getting married right away, and the vast majority of both end up getting married – though some much later than they want. This presents a different communal problem with a different set of solutions. Our community can explore new ways to help these young women (and men) who are struggling to get married.


It is this author’s opinion that contrary to Dr. Sokol’s view, the four observations do have validity and that the vast majority of women who do end up eventually marrying which Dr. Sokol references is still 2% less than his statistics of men who are married (98%).  Dr. Sokol’s figure of 96% may, unfortunately. be higher than the true number on account of the four observations.  It is also likely that reducing the dating age gap will result in more of our young women getting married – and this is a conclusion that Dr. Sokol, unfortunately, disagrees with. Perhaps a thought experiment may be employed to explain why it would be true.  What if the age gap was, say 30 years, instead of 4 years?  Would Dr. Sokol agree that, in theory, that this age gap would cause more women not to be married?  It is unlikely that he would disagree.  If so, at what point does it stop?

One last thought:  There is no question that even according to Dr. Sokol’s numbers the extra 60 percent longer in time that it takes our young ladies to get married is something that we should be addressing.


The author can be reached at [email protected]







  1. Difficult to follow an in- depth analysis of a study that has not been summarised in this article. It would be better if you would please first tell us what he said, before explaining why he is wrong…

  2. Thank you Rabbi Hoffman for elucidating what the majority of Frum people know all too well. This study has no validity at all because the data collection was woefully inadequate. At this point, the study is nothing more than a slap in the face to the thousands of unmarried frum girls & their families. All in the name of academic “research”.

  3. The larger question is, why was Dr. Sokol the first to examine the data? A hypothesis should be formed based on a close review of the data. Instead, we have organizations, as well as the author, that push solutions to problems they don’t fully understand based on empirical statistics. Imagine how much damage could have been caused by age gap reduction programs, if Dr. Sokol is right.


    Hashem already setup your sons and everyone else’s son/daughter zivug 40 days before they were even born. Forget about the fact that now they are holding at around 20 or so years later since they were born. Hashem has their zivug waiting for them and can send it to your son/daughter immediately but is just waiting for each person involved in shidduchim to do their RUCHNIUS hishtadlus of Bitachon and Emunah(faith and trust in Hashem) together with Tefillos/Prayers.

    Are YOU doing your RUCHNIUS hishtadlus? Not just your gashmius hishtadlus of dating and speaking to shadchanim etc….?

    Start turning directly to Hashem for help in all your needs, especially shidduchim that they say is a bigger miracle than the splitting of the Yam suf.

  5. We dont have a shidduch crisis. We have a “dating” (for lack of a better term) crisis. Way too many young men and women (shadchanim I have spoken to and my own personal observation, say mostly women) say no to a prospective shidduch after a single date (meeting if you would prefer). Suddenly everyone has a level of nevua we didnt know they had. Just last week, a girl said after one date “I had a wonderful time, he showed exemplary middos, he was funny and attentive too, but, I dont think he is for me”. Why? “Because I dont think so”.

    Girls more and more ask “does he have a Rav or Rebbe he is close with?”. An excellent trait to look for in a prospective mate. Yet, these very same girls turn around and using their new found powers of nevua make these decisions. Where is THEIR torah guidance? Where are their parents?

  6. Most social science attempts to study the frum community, even when done by frum scholars, are inherently flawed. First, it is a “fool’s errand” to try to define such terms as Hasidic, Yeshivish, Modern, Zionist, Baal ha-Bayis, Ben Torah, etc. with any precision. How do you class a person who wears a streimel, or one who wears a kippah srugah, who works for the goyim but spends most of his free time learning Torah? What about a yeshivish person who closely follows goyish “stuss” such as American politics, or baseball, or the stock market?

    Then you have a problem that many people being surveyed tend to tell the person conducting the survey what the respondent things the person surveying wants to hear. Then you have the problem that frum Jews, like many other people, are unlikely to admit to doing something that they presume is disapproved of (to pick a painless way, how many people admit that they think about something unrelated to davening or learning, while in shul). To pick a non-Jewish example, in the 20th century, one ethnic/national group was assumed to be at high risk for AIDS (a sexually transmitted disease) since when researchers ask patients from that group if they ever engaged in various forms of improper behavior they all denied it so the “scientists” concluded that belonging to that ethnic group was factor in contracting the disease.

    Frum social sciences research would be fun, but probably impossible to do.

  7. Out of curiosity. How many women (percentage wise) would have to be married in each group for the pattern to be considered “normal” and not a “crises”?
    Clearly it is unrealistic that every girl gets married at 20 or exactly to the first guy she goes out with. Obviously, just like some people have hardships with health and parnassoh, some people have hardships in dating.
    Is it even possible to create a statistical spread of normal? Can we go back to our parents and say their statistics is what is normal and any deviation from that is a crises? Because if we assume that previous generations were normal, then we have to accept that just like they had individuals that never got married and got married latter, so we need to accept that in this generation also.